CN112400515B

CN112400515B - Plant growth environment control method, device, equipment and storage medium based on artificial intelligence

Info

Publication number: CN112400515B
Application number: CN202011306328.8A
Authority: CN
Inventors: 曹小䶮; 罗迪君
Original assignee: Tencent Technology Shenzhen Co Ltd
Current assignee: Tencent Technology Shenzhen Co Ltd
Priority date: 2020-11-20
Filing date: 2020-11-20
Publication date: 2023-11-03
Anticipated expiration: 2040-11-20
Also published as: CN112400515A

Abstract

The application provides a plant growth environment control method and device based on artificial intelligence, electronic equipment and a computer readable storage medium; relates to artificial intelligence technology; the method comprises the following steps: encoding a plurality of environment variables applied to a plant growing environment in a first period, and constructing a first environment set by a plurality of first environment individuals obtained through encoding; reconstructing a first environmental individual in the first environmental set based on weather conditions in the first period and environmental variables in the second period to obtain a second environmental set comprising a plurality of second environmental individuals, wherein the first period is later than the second period; and determining a second environment individual with highest fitness from the second environment set, and applying a plurality of environment variables corresponding to the second environment individual with highest fitness to the plant growth environment in a second period. The application can dynamically optimize the environment variable of the plant growth environment and realize intelligent plant growth environment control.

Description

Plant growth environment control method, device, equipment and storage medium based on artificial intelligence

Technical Field

The present application relates to artificial intelligence technology, and in particular, to a plant growth environment control method, device, electronic apparatus and computer readable storage medium based on artificial intelligence.

Background

Artificial intelligence (Artificial Intelligence, AI) is a comprehensive technology of computer science, and by researching the design principles and implementation methods of various intelligent machines, the machines have the functions of sensing, reasoning and decision. Artificial intelligence technology is a comprehensive subject, and relates to a wide range of fields, such as natural language processing technology, machine learning/deep learning and other directions, and with the development of technology, the artificial intelligence technology will be applied in more fields and has an increasingly important value.

In the related art, based on the experience of planting expert, the growth environment of the plant is adjusted in a manual regulation mode so as to control the growth state of the plant, and an effective scheme for regulating the growth environment of the plant based on artificial intelligence is lacking.

Disclosure of Invention

The embodiment of the application provides a plant growth environment control method, a plant growth environment control device, electronic equipment and a computer readable storage medium based on artificial intelligence, which can dynamically optimize environment variables of a plant growth environment and realize intelligent plant growth environment control.

The technical scheme of the embodiment of the application is realized as follows:

the embodiment of the application provides a plant growth environment control method based on artificial intelligence, which comprises the following steps:

encoding a plurality of environment variables applied to a plant growing environment in a first period, and constructing a first environment set by a plurality of first environment individuals obtained through the encoding;

reconstructing a first environmental individual in the first environmental set based on the weather condition of the first period and the environmental variable of the second period to obtain a second environmental set comprising a plurality of second environmental individuals, wherein the first period is later than the second period;

and determining a second environment individual with highest fitness from the second environment set, and applying a plurality of environment variables corresponding to the second environment individual with highest fitness to the plant growth environment in the second period.

The embodiment of the application provides a plant growth environment control device, which comprises:

the construction module is used for carrying out coding processing on a plurality of environment variables applied to the plant growing environment in a first period, and constructing a first environment set through a plurality of first environment individuals obtained through the coding processing;

The reconstruction module is used for carrying out reconstruction processing on a first environmental individual in the first environmental set based on the weather condition of the first period and the environmental variable of the second period to obtain a second environmental set comprising a plurality of second environmental individuals, wherein the first period is later than the second period;

and the determining module is used for determining a second environment individual with highest fitness from the second environment set, and applying a plurality of environment variables corresponding to the second environment individual with highest fitness to the plant growth environment in the second period.

In the above technical solution, the construction module is further configured to encode a plurality of environmental variables applied to a plant growing environment in a first period, to obtain a first environmental individual including the plurality of environmental variables;

and carrying out replication processing on the first environmental individuals, and taking a set of a plurality of environmental individuals obtained by the replication processing as a first environmental set.

In the above technical solution, the first period includes a plurality of time periods; the building module is further configured to perform the following processing for any one of the plurality of environment variables:

acquiring a variable value of the environment variable in any one of the time periods;

Performing coding processing on the variable values in the time periods to obtain coded values in the time periods;

based on the sequence of the time periods, performing splicing processing on the coded values in the time periods to obtain coded data of the environment variable in the first period;

and splicing the encoded data of the plurality of environment variables in the first period to obtain a first environment individual comprising the plurality of environment variables.

In the above technical solution, the reconstruction module is further configured to iteratively execute the following processes:

determining the fitness of each first environmental individual based on the environmental variable of the second period, the weather condition of the first period and a plurality of environmental variables corresponding to the first environmental individual;

screening a plurality of first environmental individuals in the first environmental set based on the fitness of each first environmental individual, and taking the set of environmental individuals obtained after the screening as a sub-environmental set of the first environmental set;

performing conversion processing on the environment individuals in the sub-environment set to obtain a converted sub-environment set comprising a plurality of third environment individuals;

Screening the first environmental individuals in the first environmental set and the third environmental individuals in the converted sub-environmental set to obtain a second environmental set comprising a plurality of second environmental individuals, and taking the second environmental set as a new first environmental set;

and stopping the iterative processing when the iteration termination condition is met.

In the above technical solution, the reconstruction module is further configured to invoke a plant simulator model based on the weather condition of the first period, so as to determine resource information required to be consumed for converting the environmental variable of the second period into a plurality of environmental variables corresponding to the first environmental individual;

invoking the simulator model based on a plurality of environment variables corresponding to the first environment individuals to determine growth expectation information brought by the plant growth;

and taking the difference value of the profit and the resource information as the fitness of the first environmental individual.

In the above technical solution, the reconstruction module is further configured to determine fitness of each first environmental individual in the first environmental set and fitness of each third environmental individual in the converted sub-environmental set;

And based on the fitness of the environmental individuals, ordering the first environmental individuals in the first environmental set and the third environmental individuals in the sub-environmental set after the conversion processing in a descending order, and taking the set of the plurality of environmental individuals in the previous descending order ordering result as a second environmental set comprising a plurality of second environmental individuals.

In the above technical solution, the reconstruction module is further configured to determine a sampling probability of each first environmental individual based on an fitness of each first environmental individual;

and based on the sampling probability of each first environmental individual, sampling a plurality of first environmental individuals in the first environmental set, and taking the set of the first environmental individuals obtained after the sampling as a sub-environmental set of the first environmental set.

In the above technical solution, the reconstruction module is further configured to perform, for any one of the plurality of first environmental individuals in the first environmental set, the following processing:

adding the fitness of each first environmental individual to obtain a summation result;

and taking the ratio of the fitness of the first environmental individual to the addition result as the sampling probability of the first environmental individual.

In the above technical solution, the reconstruction module is further configured to perform a jth iteration process:

taking the summation of sampling probabilities of the first i first environmental individuals as a first summation result;

taking the sum of sampling probabilities of the first i+1 first environmental individuals as a second sum result;

when the random number generated by the jth iteration process is larger than the first addition result and smaller than or equal to the second addition result, the (i+1) th first environmental individual is used as the first environmental individual obtained after the sampling process;

stopping the iterative process when the j is equal to N;

wherein i and j are natural numbers which are increased from 1, the values of i and j are more than or equal to 1 and less than or equal to N-1, i is more than or equal to 1 and less than or equal to M-1, N is the total number of iterative processing, and M is the total number of the plurality of first environmental individuals.

In the above technical solution, the reconstruction module is further configured to perform a coded value exchange process on at least two environmental individuals in the sub-environmental set, to obtain a sub-environmental set including the exchanged environmental individuals;

performing coded value transformation processing on the environment individuals in the sub-environment set comprising the exchanged environment individuals, and taking the sub-environment set comprising the transformed environment individuals as a transformed sub-environment set comprising a plurality of third environment individuals.

In the above technical solution, the reconstruction module is further configured to perform, for any environmental individual in the sub-environmental set, the following processing:

randomly selecting the coding value of the environmental individual in any time period;

sampling the environmental individuals except the environmental individuals in the sub-environmental set to obtain sampling environmental individuals;

based on the exchange probability, carrying out exchange processing based on the coding value of the sampling environment individual in the time period and the coding value of the environment individual in the time period to obtain two exchanged environment individuals;

and taking the set of the two exchanged environment individuals and the environment individuals which are not subjected to the exchange processing as a sub-environment set comprising the exchanged environment individuals.

In the above technical solution, the reconstruction module is further configured to sample the environmental individuals in the sub-environmental set including the exchanged environmental individuals based on the first transformation probability, to obtain the environmental individuals to be transformed;

performing random transformation processing based on the coding values of the environmental individuals to be mutated in any time period based on the second mutation probability to obtain transformed environmental individuals;

And taking the transformed environment individuals and the set of environment individuals which are not subjected to the transformation processing as a sub-environment set which comprises a plurality of third environment individuals after the transformation processing.

An embodiment of the present application provides an electronic device for plant growth environment control, including:

a memory for storing executable instructions;

and the processor is used for realizing the plant growth environment control method based on artificial intelligence when executing the executable instructions stored in the memory.

The embodiment of the application provides a computer readable storage medium which stores executable instructions for causing a processor to execute, thereby realizing the plant growth environment control method based on artificial intelligence.

The embodiment of the application has the following beneficial effects:

the first environmental individuals in the first environmental set are rebuilt, and the environmental variables corresponding to the second environmental individuals with the highest fitness are determined from the second environmental set, so that the environmental variables of the plant growth environment are dynamically optimized, the plant growth environment is automatically controlled, and further intelligent and accurate control of the plant growth environment is realized.

Drawings

Fig. 1 is a schematic diagram of an application scenario of a control system provided by an embodiment of the present application;

FIG. 2 is a schematic diagram of an electronic device for plant growth environmental control according to an embodiment of the present application;

FIGS. 3-5 are schematic flow diagrams of an artificial intelligence-based plant growth environment control method according to an embodiment of the present application;

FIG. 6 is a schematic flow chart of an artificial intelligence-based plant growth environment control method according to an embodiment of the present application;

FIG. 7 is a schematic diagram of temperature variables provided by an embodiment of the present application;

FIG. 8 is a schematic illustration of an environmental individual provided by an embodiment of the present application;

FIG. 9A is a schematic diagram of an individual environment without exchange processing provided by an embodiment of the present application;

FIG. 9B is a schematic diagram of an individual environment subject to the exchange process provided by an embodiment of the present application;

FIG. 10A is a schematic diagram of an individual environment without transformation processing provided by an embodiment of the present application;

FIG. 10B is a schematic diagram of an individual environment subjected to a transformation process according to an embodiment of the present application;

fig. 11 is a schematic flow chart of policy iteration provided in an embodiment of the present application.

Detailed Description

The present application will be further described in detail with reference to the accompanying drawings, for the purpose of making the objects, technical solutions and advantages of the present application more apparent, and the described embodiments should not be construed as limiting the present application, and all other embodiments obtained by those skilled in the art without making any inventive effort are within the scope of the present application.

In the following description, the terms "first", "second", and the like are merely used to distinguish between similar objects and do not represent a particular ordering of the objects, it being understood that the "first", "second", or the like may be interchanged with one another, if permitted, to enable embodiments of the application described herein to be practiced otherwise than as illustrated or described herein.

Unless defined otherwise, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of the application only and is not intended to be limiting of the application.

Before describing embodiments of the present application in further detail, the terms and terminology involved in the embodiments of the present application will be described, and the terms and terminology involved in the embodiments of the present application will be used in the following explanation.

1) Convolutional neural network (CNN, convolutional Neural Networks): one type of feed-forward neural network (FNN, feedforward Neural Networks) that includes convolution calculations and has a deep structure is one of the representative algorithms of deep learning. Convolutional neural networks have the capability of token learning (representation learning) to enable a shift-invariant classification (shift-invariant classification) of input images in their hierarchical structure.

2) Environmental variable: for influencing plant growth environmental factors such as greenhouse temperature, light supplement lamps, greenhouse air humidity, greenhouse carbon dioxide concentration, etc.

3) Environmental individuals: a collection of several identical or non-identical environmental variables may change with respect to an environmental variable as a result of a change in that environmental variable, then the collection is an individual with respect to the environmental variable, i.e., an environmental individual includes a sequence or collection of various environmental variables.

The embodiment of the application provides a plant growth environment control method and device based on artificial intelligence, electronic equipment and a computer readable storage medium, which can dynamically optimize environment variables of plant growth environment and realize automatic plant growth environment control.

The plant growth environment control method based on artificial intelligence provided by the embodiment of the application can be independently realized by a terminal/server; the method may be implemented cooperatively by a terminal and a server, for example, the terminal alone bears an artificial intelligence-based plant growth environment control method described below, and applies a plurality of environment variables corresponding to a second environment individual having highest fitness to a plant growth environment in a second period through a control device, or the terminal sends an update request for the environment variables (including weather conditions of the first period and the environment variables of the second period) to the server, the server executes the artificial intelligence-based plant growth environment control method according to the received update request for the environment variables, and applies a plurality of environment variables corresponding to the second environment individual having highest fitness to the plant growth environment in the second period through the control device in response to the update request for the environment variables, so as to automatically control the growth of plants.

The electronic device for controlling the plant growth environment provided by the embodiment of the application can be various types of terminal devices or servers, wherein the servers can be independent physical servers, can be a server cluster or a distributed system formed by a plurality of physical servers, and can be cloud servers for providing cloud computing services; the terminal may be, but is not limited to, a smart phone, a tablet computer, a notebook computer, a desktop computer, a smart speaker, a smart watch, etc. The terminal and the server may be directly or indirectly connected through wired or wireless communication, and the present application is not limited herein.

Taking a server as an example, for example, a server cluster deployed in a cloud may be used, an artificial intelligence cloud Service (aias a Service, AIaaS) is opened to users, an AIaaS platform splits several common AI services and provides independent or packaged services in the cloud, and the Service mode is similar to an AI theme mall, and all users can access one or more artificial intelligence services provided by using the AIaaS platform through an application programming interface.

For example, one of the artificial intelligence cloud services may be a plant growth environment control service, that is, a cloud server is packaged with a program for controlling plant growth environment provided by the embodiment of the present application. The method comprises the steps that a user invokes a plant growth environment control service in cloud service through a terminal (a client is operated, such as a greenhouse monitoring client and the like), so that a server deployed in a cloud end invokes a program for controlling the packaged plant growth environment, a first environment individual in a first environment set is rebuilt to obtain a second environment set comprising a plurality of second environment individuals, a second environment individual with highest fitness is determined from the second environment set, a plurality of environment variables corresponding to the second environment individual with highest fitness are applied to the plant growth environment in a second period to monitor the growth of plants, for example, for a greenhouse monitoring application, a weather sensor (such as a temperature sensor and a humidity sensor and the like) in a greenhouse is used for acquiring a weather forecast in a current period, a weather forecast in the current period is used for acquiring the weather forecast in the current period, the first environment individual in the first environment set is rebuilt based on the weather forecast in the current period to obtain the second environment set comprising the plurality of second environment individuals, the second environment individual with highest fitness is determined from the second environment set, a plurality of environment variables corresponding to the second environment individual with highest fitness is applied to the plant growth in the second environment set in the second period, for example, a humidifying device (such as a device and a device is used for controlling the plant growth in the current environment in the current period) is automatically applied to the plant growth in the greenhouse, and the current environment is controlled in the current period.

Referring to fig. 1, fig. 1 is a schematic diagram of an application scenario of a control system 10 according to an embodiment of the present application, a terminal 200 is connected to a server 100 through a network 300, where the network 300 may be a wide area network or a local area network, or a combination of the two.

The terminal 200 (running with a client, such as a greenhouse monitoring client, etc.) may be used to obtain an update request for an environment variable, for example, when a preset point in time is reached, i.e. a first period, the terminal 200 obtains a weather state (weather forecast or weather data of the past year) of the first period through a third application, the terminal 200 obtains an environment variable of a second period (history period) through a sensor in the greenhouse, and after the terminal 200 collects the weather state of the first period and the environment variable of the second period, an update request for the environment variable is automatically generated.

In some embodiments, a plant growth environment control plug-in may be implanted in a client running in the terminal to implement an artificial intelligence based plant growth environment control method locally at the client. For example, after acquiring an update request for an environmental variable (including a weather condition in a first period and an environmental variable in a second period), the terminal 200 invokes the plant growth environmental control plug-in to implement an artificial intelligence-based plant growth environmental control method, reconstructs a first environmental individual in a first environmental set to obtain a second environmental set including a plurality of second environmental individuals, determines a second environmental individual with the highest fitness from the second environmental set, and applies a plurality of environmental variables corresponding to the second environmental individual with the highest fitness to a plant growth environment in the second period to automatically monitor growth of greenhouse plants.

In some embodiments, after the terminal 200 acquires the update request for the environmental variable, invokes the plant growth environmental control interface of the server 100 (may be provided in the form of cloud service, i.e. plant growth environmental control service), the server 100 reconstructs a first environmental individual in the first environmental set to obtain a second environmental set including a plurality of second environmental individuals, determines a second environmental individual with the highest fitness from the second environmental set, applies a plurality of environmental variables corresponding to the second environmental individual with the highest fitness to the plant growth environment in the second period to monitor the growth of plants, for example, for a greenhouse monitoring application, the terminal 200 acquires an environmental variable in a historical period through a sensor in a greenhouse, acquires a weather forecast in a current period through a third application, automatically generates an update request for the environmental variable based on the environmental variable in the historical period and the weather forecast in the current period, and sends the update request for the environmental variable to the server 100, the server 100 reconstructs the first environmental individual in the first environmental set to obtain a second environmental set including a plurality of second environmental individuals, determines a plurality of environmental variables corresponding to the second environmental individual with the highest fitness from the second environmental set, and sends the second environmental variable with the highest fitness to the corresponding plant growth individual to the terminal 200 in the current environmental variable to the highest fitness to the terminal 200.

Referring to fig. 2, fig. 2 is a schematic structural diagram of an electronic device 500 for controlling plant growth environment according to an embodiment of the present application, and the electronic device 500 for processing an organic matter sample shown in fig. 2 includes: at least one processor 510, a memory 550, at least one network interface 520, and a user interface 530. The various components in electronic device 500 are coupled together by bus system 540. It is appreciated that the bus system 540 is used to enable connected communications between these components. The bus system 540 includes a power bus, a control bus, and a status signal bus in addition to the data bus. The various buses are labeled as bus system 540 in fig. 2 for clarity of illustration.

The processor 510 may be an integrated circuit chip with signal processing capabilities such as a general purpose processor, such as a microprocessor or any conventional processor, or the like, a digital signal processor (DSP, digital Signal Processor), or other programmable logic device, discrete gate or transistor logic, discrete hardware components, or the like.

Memory 550 includes volatile memory or nonvolatile memory, and may also include both volatile and nonvolatile memory. The nonvolatile Memory may be a Read Only Memory (ROM), and the volatile Memory may be a random access Memory (RAM, random Access Memory). The memory 550 described in embodiments of the present application is intended to comprise any suitable type of memory. Memory 550 may optionally include one or more storage devices physically located remote from processor 510.

In some embodiments, memory 550 is capable of storing data to support various operations, examples of which include programs, modules and data structures, or subsets or supersets thereof, as exemplified below.

An operating system 551 including system programs for handling various basic system services and performing hardware-related tasks, such as a framework layer, a core library layer, a driver layer, etc., for implementing various basic services and handling hardware-based tasks;

network communication module 552 is used to reach other computing devices via one or more (wired or wireless) network interfaces 520, exemplary network interfaces 520 include: bluetooth, wireless compatibility authentication (WiFi), and universal serial bus (USB, universal Serial Bus), etc.;

In some embodiments, the plant growth environment control device provided by the embodiments of the present application may be implemented in a software manner, for example, may be a plant growth environment control plug-in the terminal described above, and may be a plant growth environment control service in the server described above. Of course, the plant growth environment control device provided by the embodiment of the present application may be provided in various forms including application programs, software modules, scripts or codes, as various software embodiments.

Fig. 2 shows a plant growing environment control device 555 stored in a memory 550, which may be software in the form of programs and plug-ins or the like, such as a plant growing environment control plug-in, and comprising a series of modules including a construction module 5551, a reconstruction module 5552 and a determination module 5553; the construction module 5551, the reconstruction module 5552 and the determination module 5553 are configured to implement the plant growth environment control function provided by the embodiment of the present application.

As described above, the plant growth environment control method based on artificial intelligence provided by the embodiment of the application can be implemented by various types of electronic devices. Referring to fig. 3, fig. 3 is a schematic flow chart of a plant growth environment control method based on artificial intelligence according to an embodiment of the present application, and is described with reference to the steps shown in fig. 3.

In the following step, the weather condition of the first period may be a weather forecast obtained by the third application, or may be weather data of the same period in the past, that is, the weather condition of the first period is a predicted weather condition of the first period and is not a real weather condition. Wherein the second period is a historical period relative to the first period.

Wherein the environmental variable of the second period is perceived by the sensor as a growing environment in the greenhouse. The environmental variable for the first period may be an initial variable set based on a priori knowledge, or may be a reference variable in a reference environmental control strategy.

In step 101, a plurality of environment variables applied to a plant growing environment during a first period are encoded, and a plurality of first environment individuals obtained by the encoding process construct a first environment set.

As shown in fig. 6, the terminal acquires environmental variables of a history period (second period) with respect to a current period (first period) through a sensor in a greenhouse, acquires weather conditions of the first period through a third application, automatically generates an update request for the environmental variables based on the environmental variables of the second period and the weather conditions of the first period, and transmits the update request for the environmental variables to a server, and after receiving the update request for the environmental variables, the server acquires a plurality of environmental variables of the first period applied to a plant growing environment, and performs encoding processing on the plurality of environmental variables of the first period applied to the plant growing environment, so as to obtain a plurality of first environmental individuals, and constructs a first environmental set based on the plurality of first environmental individuals, so as to perform dynamic optimization of the environmental variables based on the first environmental set later.

Referring to fig. 4, fig. 4 is a schematic flow chart of an alternative plant growth environment control method based on artificial intelligence according to an embodiment of the present application, and fig. 4 shows that step 101 in fig. 3 may be implemented through steps 1011 to 1012: in step 1011, encoding a plurality of environmental variables applied to a plant growing environment during a first period to obtain a first environmental individual including the plurality of environmental variables; in step 1012, the first environmental individual is subjected to a replication process, and a set of a plurality of environmental individuals obtained by the replication process is used as a first environmental set.

For example, the plurality of environmental variables of the first period applied to the plant growing environment (i.e., the plurality of environmental variables of the first period) are non-optimized variables, and the plurality of environmental variables of the first period need to be dynamically optimized to obtain the environmental variables more suitable for the plant growing environment of the first period.

The plurality of environment variables in the first period may be text-type variables, such as temperature 28 degrees celsius, humidity 80%, etc., and the text-type variables need to be encoded to obtain a processing format suitable for subsequent reconstruction. Thus, the plurality of environment variables of the first period are encoded to obtain a first environment individual comprising the plurality of environment variables, the first environment individual is duplicated, and the duplicated set of the plurality of environment individuals is used as a first environment set to form an initial environment set so as to reconstruct environment individuals in the initial environment set later.

In some embodiments, the first period includes a plurality of time periods; encoding a plurality of environmental variables applied to a plant growing environment during a first period to obtain a first environmental individual including the plurality of environmental variables, comprising: the following is performed for any one of a plurality of environment variables: acquiring variable values of the environment variable in any one of a plurality of time periods; performing coding processing on variable values in a plurality of time periods to obtain coded values in the plurality of time periods; based on the sequence of the time periods, splicing the coded values in the time periods to obtain coded data of the environment variable in the first period; and splicing the encoded data of the plurality of environment variables in the first period to obtain a first environment individual comprising the plurality of environment variables.

The first period is a period of a plant growing period, for example, the plant growing period is 70 days, and the first period may be any one day or any several days out of 70 days. The environmental variable may be temperature, humidity, carbon dioxide concentration, etc.

For example, as shown in fig. 7, the first period is one day in the plant growing period, the first period is divided into 4 periods, namely 6 hours as one period, the environmental variables are a temperature variable, a humidity variable and a carbon dioxide concentration, the temperature variable is taken as an example, variable values (temperature values) of the temperature variable in the 4 periods are obtained, for example, a first period (25 ℃), a second period (26 ℃), a third period (27 ℃) and a fourth period (29 ℃), the variable values of the temperature variable in the 4 periods are encoded, the encoded values of the temperature variable in the 4 periods are obtained, for example, the first period (11001), the second period (11010), the third period (11011) and the fourth period (11101), and the encoded values of the temperature variable in the 4 periods are spliced, so that encoded data of the temperature variable in the first period, namely [11001 11010 11011 11101], can be controlled to achieve finer granularity of the environmental variable, and even achieve a small-scale.

For example, as shown in fig. 8, when the encoded data of the temperature variable in the first period is [11001 11010 11011 11101], the encoded data of the humidity variable in the first period is [1010000 1010000 1010000 1010011], and the encoded data of the carbon dioxide concentration variable in the first period is [100101100 100101100 100101100 100101100], the encoded data of the temperature variable, the humidity variable, and the carbon dioxide concentration variable in the first period are spliced to obtain an environmental individual including a plurality of environmental variables.

In step 102, a reconstruction process is performed on a first environmental individual in the first environmental set based on the weather condition of the first period and the environmental variable of the second period, to obtain a second environmental set including a plurality of second environmental individuals, wherein the first period is later than the second period.

After the server obtains the first environment set, a part of first environment individuals in the first environment set can be rebuilt based on weather conditions in the first period and environment variables in the second period to obtain a plurality of second environment individuals, and the second environment set is built based on the plurality of second environment individuals so as to obtain the optimal environment variables in the first period from the second environment set.

Referring to fig. 5, fig. 5 is a schematic flow chart of an artificial intelligence-based plant growth environment control method according to an embodiment of the present application, and fig. 5 shows that step 102 in fig. 3 may be implemented by steps 1021-1024: the following process is iteratively performed: in step 1021, determining fitness of each first environmental individual based on the environmental variable of the second time period, the weather condition of the first time period, and the plurality of environmental variables corresponding to the first environmental individuals; in step 1022, based on the fitness of each first environmental individual, a screening process is performed on a plurality of first environmental individuals in the first environmental set, and the set of environmental individuals obtained after the screening process is used as a sub-environmental set of the first environmental set; in step 1023, performing conversion processing on the environment individuals in the sub-environment set to obtain a converted sub-environment set including a plurality of third environment individuals; in step 1024, a screening process is performed on the first environmental individuals in the first environmental set and the third environmental individuals in the converted sub-environmental set, so as to obtain a second environmental set including a plurality of second environmental individuals, and the second environmental set is used as a new first environmental set; and stopping the iterative process when the iteration termination condition is satisfied.

The iteration termination condition may be the iteration number, the iteration time, etc., for example, stopping the iteration process when the preset iteration number is reached; stopping the iterative process when the preset iterative time is reached; stopping the iterative process when the highest fitness in the second environment set in the iteration results of T times is the same, wherein T is a natural number greater than 1.

Through the iterative processing, the environmental individuals in the environmental set can be continuously optimized to obtain the optimal environmental individuals, and the corresponding environmental variables in the optimal environmental individuals are applied to the plant growth environment in the second period so as to better control the plant growth.

In some embodiments, determining the fitness of each first environmental individual based on the environmental variable for the second time period, the weather condition for the first time period, and the plurality of environmental variables corresponding to the first environmental individual comprises: invoking a plant simulator model based on the weather conditions of the first period to determine resource information required to be consumed for converting the environmental variables of the second period to a plurality of environmental variables corresponding to the first environmental individuals; invoking a simulator model based on a plurality of environment variables corresponding to the first environment individuals to determine growth expected information brought by plant growth; and taking the difference value of the profit and the resource information as the fitness of the first environmental individual.

For example, in order to screen the environmental individuals in the environmental set by their fitness, the fitness of the environmental individuals may be obtained based on a plant simulator model, which is a neural network model, such as a convolutional neural network model, a deep neural network model, or the like. The plant simulator model is used for simulating a greenhouse environment and simulating plant growth, namely, the plant simulator determines resource information required to be consumed for converting and processing the environmental variable in the second period to a plurality of environmental variables corresponding to the first environmental individuals based on the weather condition in the first period, for example, equipment consumption when equipment is controlled to convert the environmental variable; growth expectancy information (i.e., benefits) caused by plant growth is determined based on a plurality of environmental variables corresponding to the first environmental individual, e.g., simulating plant growth during the first period may yield fruit yield, fruit quality, etc. yield information. The greenhouse environment and the plant growth are simulated through the plant simulator model, so that the advantages and disadvantages of the environment individuals are accurately reflected.

In some embodiments, based on the fitness of each first environmental individual, performing a screening process on a plurality of first environmental individuals in the first environmental set, and taking a set of environmental individuals obtained after the screening process as a sub-environmental set of the first environmental set, including: determining a sampling probability of each first environmental individual based on the fitness of each first environmental individual; based on the sampling probability of each first environmental individual, sampling a plurality of first environmental individuals in a first environmental set, and taking the set of the first environmental individuals obtained after the sampling as a sub-environmental set of the first environmental set.

And screening a plurality of first environmental individuals from the first environmental sets based on the sampling probability of each first environmental individual, and taking the screened sets of the plurality of first environmental individuals as sub-environmental sets of the first environmental sets, wherein when the sampling probability of the environmental individuals is higher, the sampling probability of the environmental individuals is higher. Wherein the sampling probability for each first environmental individual is determined by: performing the following processing for any one of a plurality of first environmental individuals in the first environmental set: adding the fitness of each first environmental individual to obtain an adding result; and taking the ratio of the fitness of the first environmental individual to the addition result as the sampling probability of the first environmental individual.

The specific sampling process is as follows: performing the j-th iteration process: taking the summation of sampling probabilities of the first i first environmental individuals as a first summation result; taking the sum of sampling probabilities of the first i+1 first environmental individuals as a second sum result; when the random number generated by the jth iteration process is larger than the first addition result and smaller than or equal to the second addition result, taking the (i+1) th first environmental individual as a first environmental individual obtained after sampling processing; stopping the iterative process when j is equal to N; wherein i and j are natural numbers which are increased from 1, the values of i and j are more than or equal to 1 and less than or equal to N-1, i is more than or equal to 1 and less than or equal to M-1, N is the total number of iterative processing, and M is the total number of a plurality of first environmental individuals.

For example, the sum of sampling probabilities of the first M first environmental individuals is 1, and a random number k is generated in the jth iterative process, where 0 < k is less than or equal to 1, for example, k is 0.5, the sum of sampling probabilities of the first i first environmental individuals is 0.4, the sum of sampling probabilities of the first i+1 first environmental individuals is 0.6, the (i+1) th first environmental individual is taken as the first environmental individual obtained after the sampling process, and the operation is repeated N times to obtain the first environmental individual obtained after the N sampling processes, where the first environmental individuals obtained after the N sampling processes may be repeated.

In some embodiments, performing conversion processing on the environment individuals in the sub-environment set to obtain a converted sub-environment set including a plurality of third environment individuals, including: performing coded value exchange processing on at least two environment individuals in the sub-environment set to obtain a sub-environment set comprising exchanged environment individuals; performing coded value transformation processing on the environment individuals in the sub-environment set comprising the exchanged environment individuals, and taking the sub-environment set comprising the transformed environment individuals as a transformed sub-environment set comprising a plurality of third environment individuals.

Wherein the coded values of two environmental individuals can be changed in form by the coded value exchange process, and the coded value of a certain environmental individual can be changed substantially by the code conversion process. By combining formally changing and substantially changing the environmental individuals in the conversion sub-environment set, dynamic optimization of environmental variables is expedited.

In some embodiments, performing coded value exchange processing on at least two environmental individuals in the sub-environmental set to obtain a sub-environmental set including exchanged environmental individuals, including: the following is performed for any of the environmental individuals in the set of sub-environments: randomly selecting the coding value of the environmental individual in any time period; sampling the environmental individuals except the environmental individuals in the sub-environmental set to obtain a sampling environmental individual; based on the exchange probability, carrying out exchange processing based on the code value of the sampling environment individual in the time period and the code value of the environment individual in the time period to obtain two exchanged environment individuals; and taking the set of the two exchanged environment individuals and the environment individuals which are not subjected to the exchange processing as a sub-environment set comprising the exchanged environment individuals.

As shown in fig. 9A, the environmental individuals 1 and 2 in the sub-environmental set are arbitrarily selected, where the encoded data of the environmental individuals 1 in the first period is [11001 11001 1010000 1010000], the encoded data of the environmental individuals 2 in the first period is [11001 11001 1010011 1010011], and then the encoded value of the humidity variable in the environmental individuals 1 in the second period is exchanged with the encoded value of the humidity variable in the environmental individuals 2 in the second period, so as to obtain two environmental individuals as shown in fig. 9B, that is, in fig. 9B, the encoded data of the environmental individuals 1 in the first period is [11001 11001 1010000 1010011], and the encoded data of the environmental individuals 2 in the first period is [11001 11001 1010011 1010000].

In some embodiments, performing a coded value transformation process on an individual environment in a sub-environment set including the exchanged individual environments, taking the sub-environment set including the transformed individual environments as a transformed sub-environment set including a plurality of third individual environments, including: sampling environmental individuals in a sub-environmental set comprising exchanged environmental individuals based on the first transformation probability to obtain environmental individuals to be transformed; performing random transformation processing based on the coding value of the environmental individual to be mutated in any time period based on the second mutation probability to obtain a transformed environmental individual; and taking the set of the transformed environment individuals and the environment individuals which are not subjected to transformation processing as a sub-environment set which comprises a plurality of third environment individuals after transformation processing.

As shown in fig. 10A, based on the first transformation probability, one environmental individual 3 (environmental individual to be transformed) in the sub-environmental set is selected, the encoded data of the environmental individual 3 in the first period is [11001 11001 1010000 1010000], and based on the second transformation probability, the encoded value (11001) of the temperature variable in the environmental individual 3 in the first period is selected for transformation, wherein the value range of the temperature variable is 25-29 ℃, the encoded value (11001) can be randomly transformed into the encoded value (11010) or the encoded value (11011) or the encoded value (11100) or the encoded value (11101), and as shown in fig. 10B, the encoded data of the environmental individual 3 in the first period is [11010 11001 1010000 1010000].

In some embodiments, filtering the first environmental individuals in the first environmental set and the third environmental individuals in the converted sub-environmental set to obtain a second environmental set including a plurality of second environmental individuals includes: determining the fitness of each first environmental individual in the first environmental set and the fitness of each third environmental individual in the sub-environmental set after conversion processing; and based on the fitness of the environmental individuals, carrying out descending order sequencing on the first environmental individuals in the first environmental set and the third environmental individuals in the sub-environmental set after the conversion processing, and taking the set of the plurality of environmental individuals in the previous descending order sequencing result as a second environmental set comprising a plurality of second environmental individuals.

The method comprises the steps of screening environment individuals in a first environment set and a sub-environment set after conversion processing to obtain optimal environment individuals in the first environment set and the sub-environment set after conversion processing so as to continuously optimize iteration.

In step 103, a second environmental individual with the highest fitness is determined from the second environmental set, and a plurality of environmental variables corresponding to the second environmental individual with the highest fitness are applied to the plant growing environment in a second period.

As shown in fig. 6, after determining a plurality of environment variables corresponding to the second environmental individuals with the highest fitness from the second environment set, the server sends the plurality of environment variables corresponding to the second environmental individuals with the highest fitness to the terminal, and the terminal triggers the control device to regulate the plant growth environment based on the plurality of environment variables corresponding to the second environmental individuals with the highest fitness, so that the plant growth environment reaches the environmental index of the plurality of environment variables corresponding to the second environmental individuals with the highest fitness, for example, the carbon dioxide concentration reaches 300 g/m, the temperature reaches 28 ℃, and the humidity reaches 80%.

In the following, an exemplary application of the embodiments of the present application in an actual crop planting application scenario will be described.

The embodiment of the application can be applied to the application scene of crop planting, dynamically optimizes the control decision of the growth environment of crops in an automatic greenhouse, realizes fine control, can obtain higher crop yield with less labor cost and resource consumption compared with a manual control mode, and has popularization. For example, the following scenarios may be specifically applied: 1) The environment decision planning of actual planting is carried out, and before the actual planting, the environment decision planning is carried out in a crop simulator rapidly to obtain a better environment control decision for the whole period; 2) Dynamically optimizing an actual planting environment control decision, and in the actual planting process, iteratively optimizing a future environment control decision according to environmental observation data in a greenhouse and a future weather forecast; 3) The environmental control decision is quickly popularized to different planting areas, the historical weather data of the designated area is input into the crop simulator, and the environmental control decision suitable for the designated area can be directly and quickly iterated through the crop simulator.

In the related art, most of environment decision schemes for planting crops in an automatic greenhouse are planting expert experience and manual regulation. Before planting is started, environmental control decisions of the whole planting period are made according to the climate, the growth period, the planting variety, environmental control equipment and the like of a specific planting area. In the planting process, the environment control decision is adjusted by adopting a manual regulation mode according to the growth condition of crops.

However, the environmental decision schemes of automated greenhouse crop planting in the related art have the following problems:

1) The environmental decision scheme based on expert experience is static in the whole planting period, and the external weather in the planting process is dynamic, so that the actual control effect of the environmental control equipment in the greenhouse can be influenced, the deviation between the crop growth state and the expected crop is further expanded, and a large gap exists between the final actual planting effect and the expected crop.

2) The environment decision scheme based on expert experience completely depends on planting experience of a strategy maker, the environment decision dimension comprises temperature, illumination, irrigation, carbon dioxide concentration, humidity, fertilizer and the like, the whole planting period of various crops is usually not less than 100 days, and the decision space is quite large. Meanwhile, the planting experience needs to be accumulated by observing the growth state of crops after decision making, so that the factors influencing the growth condition of the crops are numerous, and the state space is huge. In addition, crop planting is aimed at final harvest, so that it is necessary to go through a complete planting cycle to accumulate a planting experience. Therefore, even an expert with many years of planting experience can only give an environmental decision scheme in a very small part of decision space and state space, and a large gap exists between the environmental decision scheme and a true global optimal decision.

3) Environmental decision schemes based on expert experience are coarse-grained, typically changing decision control every half month or even longer, and the decision dimension of individual variables per day is typically 1-3. This is because the outside weather is dynamically changing, while the environmental decision is static, and the use of fine control may lead to a larger deviation of the crop growth process from the expected one, so that only a milder decision change frequency and decision value can be given, and therefore the space for expert optimization decisions is very limited.

4) Environmental decision-making schemes based on expert experience are difficult to adapt to local conditions, and the climate conditions in different regions can vary widely, while experts are generally familiar with the climate conditions in only one region. Even if historical weather data is provided to the expert in the corresponding region, it is difficult for the non-data analysis practitioner to quickly and accurately analyze the climate differences in different regions and to migrate and adjust the corresponding environmental decisions.

In order to solve the above problems, the embodiments of the present application provide a dynamic optimization method for controlling the growth environment of greenhouse crops, which can be used to give a better crop planting strategy, i.e. achieve higher yield with lower cost and resource consumption through dynamic optimization, automation and fine control. Under the conditions of given crop types, facility environment data, climate of planting areas and growth period, a crop growth simulator can be utilized to simultaneously optimize a plurality of decision dimensions (environment variables), such as temperature control, light-on duration and time per day, carbon dioxide concentration, air humidity and the like, the decision dimensions adopted specifically can be customized according to the conditions of control equipment, and the granularity of control can reach the level of hours.

The following specifically describes a dynamic optimization method for controlling the growth environment of automatic greenhouse crops, which is provided by the embodiment of the application:

in order to provide a good growing environment for crops in an automated greenhouse, environmental control decisions need to be constantly optimized. First, an environmental control strategy of the whole planting period is searched (the whole period is searched), then the searched environmental control strategy is used as a reference, and after being deployed in a real greenhouse, the environmental control strategy of the next date is dynamically optimized in the future according to weather forecast and historical observation data every day (dynamic searching).

Wherein the optimization of the environmental control decisions is aided by a crop simulator. Functional modules of the crop simulator include greenhouse environment simulation and crop growth simulation. For greenhouse environment simulation, a target environment state is given first, and then the crop simulator simulates the operation of environment control equipment in a greenhouse in a closed-loop control mode based on the weather state of a planting area, so that the greenhouse climate in the crop simulator is close to the target environment state. For crop growth simulation, the crop simulator is able to simulate the growth process of the crop over a period of time given the greenhouse climate conditions of the period of time. Natural resources consumed by using control equipment can be calculated by simulating the greenhouse environment, and the simulation of crop growth can obtain harvest information such as fruit yield, fruit quality and the like, so that the net profit of planting can be calculated.

The following is a detailed description: 1) Searching the whole period; 2) Dynamic search:

1) Searching for the whole period

The environment variables to be controlled in the embodiment of the application include greenhouse temperature, light supplementing lamps, greenhouse relative humidity and greenhouse CO2 concentration. Wherein, the light supplement lamp has only 2 dimensions each day, the rest environment variables (greenhouse temperature, greenhouse relative humidity, greenhouse CO2 concentration) have 24 dimensions each day, and then all environment variables have 74 dimensions each day.

Wherein, the evaluation index of the environmental control strategy is the final net profit of the whole planting period, so one evaluation needs to input the strategy of each day in the specified planting period. However, the strategy of searching the whole period at a time has a high dimension of environment variables.

To reduce the search space, a day-by-day search is employed, i.e., searching is started sequentially from the first day. And initializing and fixing the strategies on the rest days when searching on the first day, namely exploring only 74 dimensions on the first day, and inputting the explored new strategy and the strategies on the rest days into a crop simulator for simulation together, and calculating net profits according to simulation results to serve as the fitness of the environmental individuals. And after the strategy iteration of the first day is completed, acquiring a first day strategy corresponding to the optimal environment individual, and taking the first day strategy as a fixed strategy of the first day in the searching process of other dates. When searching for the strategy of the next day, the date after the next day still adopts the initialization strategy and remains unchanged, and only the first day adopts the optimal strategy searched previously. In this way, an optimal strategy for the whole planting cycle will be obtained until the last day of the search.

As shown in fig. 11, the policy iteration flow for a certain day is as follows:

step 1: and (5) encoding environment variables. Any one environment variable has a plurality of dimensions in a day, wherein the value range of each dimension is the same, the value of each dimension is discretized into an integer, then according to the length of the value range, binary coding with a specific length is adopted for each dimension, namely 0-1 character string, each character string corresponds to a coding value, then the coding data of one environment variable are obtained by splicing according to time sequence, for example, the temperature variable has 24 dimensions in the day, wherein the value range of the temperature variable in 24 dimensions is [10, 20], then the coding length of the temperature variable in 24 dimensions is 5, for example, the coding value of the temperature variable in 1 dimension is 12 ℃, then the coding value of the temperature variable in 1 dimension is 01100, and the binary coding values of the temperature variable in 2-24 dimensions are respectively 5 bits, and then the coding data of the temperature variable in the day are obtained after splicing according to time sequence (1 dimension in each hour).

Step 2: the environment set is initialized. According to priori agronomic knowledge, an initial value is set for the phenotype of an environmental individual, and the coding data of the environmental individual is obtained through multi-environmental variable coding. The scale of one environment set is set as N, namely, the environment individual is duplicated N times to be used as an initial environment set. The same initial environment set is adopted in order that after the algorithm is iterated for a certain number of times and is terminated, if the algorithm is further iterated, the algorithm can directly continue iterating the more optimal solution under the current optimal condition.

Step 3: and (5) selecting. Firstly, calculating the fitness of each environmental individual, decoding each coded data into a phenotype, and taking each environmental individual as input data of the current optimization date. For decisions earlier than the current date in the planting period, the optimal decisions of each date obtained by previous iteration are adopted, and environment variables after the current date are used as decisions through agronomic priors. In addition, the input to the crop simulator also comprises contemporaneous historical weather condition data, the contemporaneous historical weather condition data and the environmental control decision are input to the crop simulator for simulation, and net profit is calculated according to the simulation result and is taken as the fitness of an environmental individual. And selecting an environment set with a designated scale of N according to the individual fitness of the environment for the next transformation operation.

The fitness function value guides the searching direction, decides which environmental individuals remain and which individuals are screened out, and generally depends on the objective function of the optimization problem.

The criterion for evaluating the quality of the environmental individuals is the respective fitness of the environmental individuals, and the higher the fitness of the environmental individuals is, the more opportunities are selected. The selection operation is implemented in a number of ways, for example, the probability of each individual environment being selected can be calculated first f(S _i ) Indicating the fitness of the environmental individual i; then randomly generating a random number q ₀ If (3)The environment individual i +1 is selected into a new environment set for the next transformation operation. />

Step 4: and (5) exchanging. The method comprises the steps of adopting a two-point exchange method to exchange coded data, namely, firstly randomly selecting the same positions of two environment individuals in a sub-environment set, then pairing the selected environment individuals pairwise according to exchange probability Pc, randomly setting two exchange points in coded data corresponding to the two environment individuals paired with each other, and then exchanging coded values of the two environment individuals between the two exchange points. This process reflects a random information exchange with the aim of creating new environmental individuals. The switching may be performed as a single point switch or a multi-point switch.

Step 5: and (5) transforming. A binary transformation method is adopted, and a certain coding value in the coding data in the sub-environment set is changed with equal probability. The transformation operation is performed by bits (bit), i.e. the individual environment is randomly selected from the environment set according to the transformation probability Pm, and then some coding values of the individual environment are randomly changed, i.e. 1 is changed to 0, and 0 is changed to 1. When all environmental individuals are identical in phenotype, the exchange is unable to generate new environmental individuals, and then new environmental individuals can only be generated by transformation. That is, the transformation adds a globally optimized trait.

Step 6: and (5) preferentially selecting. The fitness of each environmental individual in the sub-environmental set is calculated by the crop simulator. And sequencing the environmental individuals of the parent environmental set and the child environmental set according to the fitness, and preferentially selecting the environmental individuals with the scale of N as the next environmental set.

Step 7: and judging that the iteration is ended. If the iteration termination condition is met, taking the phenotype corresponding to the environmental individual with the highest current fitness as a final decision output; otherwise, jumping to the step 3, and continuing iteration.

And taking the output of the decision iteration of each date as a reference environmental control strategy of the date, taking the reference environmental control strategy as the priori of an initialized environmental set along with the change of the planting date, and continuing iterating a new environmental control strategy in a dynamic searching mode.

2) Dynamic searching

In an automated greenhouse, the state of each environmental variable is collected as historical observations using sensors. And acquiring the historical weather of the planting from the beginning to the moment and the weather state of the next date through weather forecast. At the current date, only the environment variable of the next date is searched, and the searching mode is similar to the process. The difference is that the dynamic search method is to replace the environmental control decision before the current date with actual historical observation data, replace the weather condition before the current date with actual historical weather condition, and acquire the weather condition of the next date by adopting weather forecast.

For example, in an actual planting area, a planting cycle and crop variety are given, and local historical weather data is obtained. And then iterating the environment control decision of the whole planting period, and deploying the final optimal decision to the actual greenhouse in a remote control mode. By setting the time interval of control, the remote control program can automatically generate the control decision of each current environment variable at regular time and send the control decision to the control equipment of the actual greenhouse in real time through the control interface. At the same time of remote control program deployment, a remote data acquisition program is started to continuously collect and process sensor data of the actual greenhouse. After the historical sensor data is acquired, a dynamic optimizer will be enabled. Unlike pre-deployment decision optimization, dynamic optimization controls decisions only for the future 1 day environment. In addition, the historical environmental status is determined, and the historical and future weather forecast is updated. Therefore, the historical sensor data is used as the historical environmental state, the external weather data is updated at the same time, the process is used for generating a new environmental control decision, the new environmental control decision is input into the crop simulator for interactive iteration, and the environmental control decision of the next date is updated before the current day is about to end, so that the dynamic optimization of the decision is realized.

In summary, the dynamic optimization method for controlling the growth environment of the greenhouse crops automatically provided by the embodiment of the application has the following beneficial effects:

1) The dynamic optimization of the environmental control decision can be realized, after the environmental control decision is deployed to the actual control, the environmental observation data in the greenhouse can be obtained through the sensor, the external weather data is updated according to future weather forecast, and the environmental decision is iteratively optimized by utilizing the flow;

2) The dependence on manual experience can be eliminated in a plurality of environment control decision dimensions, and a relatively better strategy can be quickly searched globally. The crop simulator can iterate a complete planting period for 15 seconds on average, so that planting data can be accumulated rapidly, and then a better environment control decision can be obtained through a small number of iteration times under the condition of a larger state space according to the flow. Therefore, the iteratively obtained environmental control decisions are closer to optimal decisions than expert decisions;

3) An hourly level environmental control decision scheme may be provided because the crop simulator may simulate hourly strategies and conditions, such that environmental control decisions may be iterated at an hourly granularity, ultimately providing a finer granularity strategy than expert decisions. In addition, by adopting a dynamic optimization mode, the deviation between crop growth and simulation can be reduced to a certain extent.

4) The method can be rapidly popularized to areas with different climates, and can directly and rapidly iterate the environmental control decision suitable for the local area by only utilizing the historical weather data of the appointed area.

The artificial intelligence-based plant growth environment control method provided by the embodiment of the application has been described so far in connection with the exemplary application and implementation of the server provided by the embodiment of the application. The embodiment of the application also provides a plant growth environment control device, and in practical application, each functional module in the plant growth environment control device can be cooperatively realized by hardware resources of electronic equipment (such as terminal equipment, a server or a server cluster), computing resources such as a processor, communication resources (such as communication in various modes for supporting realization of optical cables, cells and the like) and a memory. Fig. 2 shows a plant growing environment control device 555 stored in a memory 550, which may be software in the form of programs and plug-ins, etc., for example, software modules designed in a programming language such as software C/c++, java, etc., application software designed in a programming language such as C/c++, java, etc., or dedicated software modules in a large software system, application program interfaces, plug-ins, cloud services, etc., the different implementations being exemplified below.

Example one plant growth environment control device is a mobile terminal application and module

The plant growth environment control device 555 in the embodiment of the application can be provided as a software module designed by using programming languages such as software C/C++, java and the like, and is embedded into various mobile terminal applications (stored in a storage medium of a mobile terminal as executable instructions and executed by a processor of the mobile terminal) of an Android or iOS-based system, so that related information recommendation tasks are completed by directly using the computing resources of the mobile terminal, and processing results are transmitted to a remote server in a periodic or aperiodic manner through various network communication modes or are stored locally at the mobile terminal.

Example two, plant growth environment control device is a server application and platform

The plant growth environment control device 555 in the embodiment of the application can be provided as application software designed by using programming languages such as C/C++, java and the like or a special software module in a large software system, and runs on a server side (the application software is stored in a storage medium of the server side in a mode of executable instructions and is run by a processor of the server side), and the server uses own computing resources to complete related information recommendation tasks.

The embodiment of the application can also be used for carrying a customized and easy-to-interact network (Web) Interface or other User Interfaces (UI) on a distributed and parallel computing platform formed by a plurality of servers to form an information recommendation platform (used for recommendation lists) for individuals, groups or units, and the like.

Example three plant growth Environment control device is a Server-side application program interface (API, application Program Interface) and plug-in

The plant growth environment control device 555 in the embodiment of the application can be provided as an API or plug-in on the server side for a user to call so as to execute the plant growth environment control method based on artificial intelligence in the embodiment of the application and be embedded into various application programs.

Example four the plant growth environment control device is a mobile device client API and plug-in

The plant growth environment control device 555 in the embodiment of the application can be provided as an API or a plug-in on the mobile equipment side for a user to call so as to execute the plant growth environment control method based on artificial intelligence in the embodiment of the application.

Example five the plant growth environment control device is a cloud open service

The plant growth environment control device 555 in the embodiment of the application can provide information recommendation cloud service developed for users, so that individuals, groups or units can acquire recommendation lists.

The plant growth environment control device 555 includes a series of modules, including a construction module 5551, a reconstruction module 5552, and a determination module 5553. The following continues to describe the implementation of a plant growth environment control scheme by cooperation of each module in the plant growth environment control device 555 provided by the embodiment of the present application.

A construction module 5551, configured to perform coding processing on a plurality of environment variables applied to a plant growing environment in a first period, and construct a first environment set by using a plurality of first environment individuals obtained through the coding processing; a reconstruction module 5552, configured to perform reconstruction processing on a first environmental individual in the first environmental set based on the weather condition of the first period and the environmental variable of the second period, to obtain a second environmental set including a plurality of second environmental individuals, where the first period is later than the second period; a determining module 5553, configured to determine a second environmental individual with highest fitness from the second environmental set, and apply a plurality of environmental variables corresponding to the second environmental individual with highest fitness to the plant growth environment in the second period.

In some embodiments, the constructing module 5551 is further configured to encode a plurality of environment variables applied to the plant growing environment during the first period, to obtain a first environment individual including the plurality of environment variables; and carrying out replication processing on the first environmental individuals, and taking a set of a plurality of environmental individuals obtained by the replication processing as a first environmental set.

In some embodiments, the first period includes a plurality of time periods; the building module 5551 is further configured to perform the following processing for any one of the plurality of environment variables: acquiring a variable value of the environment variable in any one of the time periods; performing coding processing on the variable values in the time periods to obtain coded values in the time periods; based on the sequence of the time periods, performing splicing processing on the coded values in the time periods to obtain coded data of the environment variable in the first period; and splicing the encoded data of the plurality of environment variables in the first period to obtain a first environment individual comprising the plurality of environment variables.

In some embodiments, the reconstruction module 5552 is further configured to iteratively perform the following: determining the fitness of each first environmental individual based on the environmental variable of the second period, the weather condition of the first period and a plurality of environmental variables corresponding to the first environmental individual; screening a plurality of first environmental individuals in the first environmental set based on the fitness of each first environmental individual, and taking the set of environmental individuals obtained after the screening as a sub-environmental set of the first environmental set; performing conversion processing on the environment individuals in the sub-environment set to obtain a converted sub-environment set comprising a plurality of third environment individuals; screening the first environmental individuals in the first environmental set and the third environmental individuals in the converted sub-environmental set to obtain a second environmental set comprising a plurality of second environmental individuals, and taking the second environmental set as a new first environmental set; and stopping the iterative processing when the iteration termination condition is met.

In some embodiments, the reconstruction module 5552 is further configured to invoke a plant simulator model based on the weather conditions of the first time period to determine resource information required to be consumed to transform the environmental variables of the second time period to the plurality of environmental variables corresponding to the first environmental individual; invoking the simulator model based on a plurality of environment variables corresponding to the first environment individuals to determine growth expectation information brought by the plant growth; and taking the difference value of the profit and the resource information as the fitness of the first environmental individual.

In some embodiments, the reconstruction module 5552 is further configured to determine a fitness of each first environmental individual in the first environmental set and a fitness of each third environmental individual in the converted sub-environmental set; and based on the fitness of the environmental individuals, ordering the first environmental individuals in the first environmental set and the third environmental individuals in the sub-environmental set after the conversion processing in a descending order, and taking the set of the plurality of environmental individuals in the previous descending order ordering result as a second environmental set comprising a plurality of second environmental individuals.

In some embodiments, the reconstruction module 5552 is further configured to determine a sampling probability for each of the first environmental individuals based on the fitness of each of the first environmental individuals; and based on the sampling probability of each first environmental individual, sampling a plurality of first environmental individuals in the first environmental set, and taking the set of the first environmental individuals obtained after the sampling as a sub-environmental set of the first environmental set.

In some embodiments, the reconstruction module 5552 is further configured to perform the following processing for any one of a plurality of first environmental individuals in the first environmental set: adding the fitness of each first environmental individual to obtain a summation result; and taking the ratio of the fitness of the first environmental individual to the addition result as the sampling probability of the first environmental individual.

In some embodiments, the reconstruction module 5552 is further configured to perform a j-th iteration process: taking the summation of sampling probabilities of the first i first environmental individuals as a first summation result; taking the sum of sampling probabilities of the first i+1 first environmental individuals as a second sum result; when the random number generated by the jth iteration process is larger than the first addition result and smaller than or equal to the second addition result, the (i+1) th first environmental individual is used as the first environmental individual obtained after the sampling process; stopping the iterative process when the j is equal to N; wherein i and j are natural numbers which are increased from 1, the values of i and j are more than or equal to 1 and less than or equal to N-1, i is more than or equal to 1 and less than or equal to M-1, N is the total number of iterative processing, and M is the total number of the plurality of first environmental individuals.

In some embodiments, the reconstruction module 5552 is further configured to perform a coded value exchange process on at least two environmental individuals in the sub-environment set, to obtain a sub-environment set including the exchanged environmental individuals; performing coded value transformation processing on the environment individuals in the sub-environment set comprising the exchanged environment individuals, and taking the sub-environment set comprising the transformed environment individuals as a transformed sub-environment set comprising a plurality of third environment individuals.

In some embodiments, the reconstruction module 5552 is further configured to perform the following processing for any environmental individual in the set of sub-environments: randomly selecting the coding value of the environmental individual in any time period; sampling the environmental individuals except the environmental individuals in the sub-environmental set to obtain sampling environmental individuals; based on the exchange probability, carrying out exchange processing based on the coding value of the sampling environment individual in the time period and the coding value of the environment individual in the time period to obtain two exchanged environment individuals; and taking the set of the two exchanged environment individuals and the environment individuals which are not subjected to the exchange processing as a sub-environment set comprising the exchanged environment individuals.

In some embodiments, the reconstruction module 5552 is further configured to sample the environmental individuals in the sub-environmental set including the exchanged environmental individuals based on the first transformation probability, to obtain environmental individuals to be transformed; performing random transformation processing based on the coding values of the environmental individuals to be mutated in any time period based on the second mutation probability to obtain transformed environmental individuals; and taking the transformed environment individuals and the set of environment individuals which are not subjected to the transformation processing as a sub-environment set which comprises a plurality of third environment individuals after the transformation processing.

Embodiments of the present application provide a computer program product or computer program comprising computer instructions stored in a computer readable storage medium. The processor of the computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the plant growth environment control method based on artificial intelligence according to the embodiment of the application.

Embodiments of the present application provide a computer readable storage medium having stored therein executable instructions which, when executed by a processor, cause the processor to perform the artificial intelligence based plant growth environment control method provided by embodiments of the present application, for example, as shown in fig. 3-5.

In some embodiments, the computer readable storage medium may be FRAM, ROM, PROM, EPROM, EEPROM, flash memory, magnetic surface memory, optical disk, or CD-ROM; but may be a variety of devices including one or any combination of the above memories.

In some embodiments, the executable instructions may be in the form of programs, software modules, scripts, or code, written in any form of programming language (including compiled or interpreted languages, or declarative or procedural languages), and they may be deployed in any form, including as stand-alone programs or as modules, components, subroutines, or other units suitable for use in a computing environment.

As an example, the executable instructions may, but need not, correspond to files in a file system, may be stored as part of a file that holds other programs or data, for example, in one or more scripts in a hypertext markup language (HTML, hyper Text Markup Language) document, in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub-programs, or portions of code).

As an example, executable instructions may be deployed to be executed on one computing device or on multiple computing devices located at one site or, alternatively, distributed across multiple sites and interconnected by a communication network.

The foregoing is merely exemplary embodiments of the present application and is not intended to limit the scope of the present application. Any modification, equivalent replacement, improvement, etc. made within the spirit and scope of the present application are included in the protection scope of the present application.

Claims

1. A plant growth environment control method based on artificial intelligence, the method comprising:

performing the following process for any one of a plurality of environmental variables applied to a plant growing environment during a first period, wherein the first period comprises a plurality of time periods;

Splicing the coded data of the plurality of environment variables in the first period to obtain a first environment individual comprising the plurality of environment variables;

copying the first environmental individuals, and taking a set of a plurality of environmental individuals obtained by the copying as a first environmental set;

2. The method of claim 1, wherein reconstructing a first environmental individual in the first environmental set based on the weather condition of the first period and the environmental variable of the second period to obtain a second environmental set including a plurality of second environmental individuals comprises:

the following process is iteratively performed:

and stopping the iterative process when the iteration termination condition is satisfied.

3. The method of claim 2, wherein the determining the fitness of each of the first environmental individuals based on the environmental variable of the second time period, the weather condition of the first time period, and the plurality of environmental variables corresponding to the first environmental individuals comprises:

Invoking a plant simulator model based on the weather conditions of the first period to determine resource information required to be consumed for converting the environmental variables of the second period to a plurality of environmental variables corresponding to the first environmental individuals;

4. The method according to claim 2, wherein the filtering the first environmental individuals in the first environmental set and the third environmental individuals in the converted sub-environmental set to obtain a second environmental set including a plurality of second environmental individuals includes:

determining the fitness of each first environmental individual in the first environmental set and the fitness of each third environmental individual in the converted sub-environmental set;

5. The method according to claim 2, wherein the screening the plurality of first environmental individuals in the first environmental set based on the fitness of each first environmental individual, and using the set of environmental individuals obtained by the screening as the sub-environmental set of the first environmental set includes:

determining sampling probability of each first environmental individual based on the fitness of each first environmental individual;

6. The method of claim 5, wherein determining the sampling probability for each of the first environmental individuals based on the fitness of each of the first environmental individuals comprises:

performing the following processing for any one of a plurality of first environmental individuals in the first environmental set:

7. The method according to claim 5, wherein the sampling the plurality of first environmental individuals in the first environmental set based on the sampling probability of each first environmental individual, and taking the set of first environmental individuals obtained after the sampling as the sub-environmental set of the first environmental set includes:

performing the j-th iteration process:

when the random number generated by the jth iteration process is larger than the first addition result and smaller than or equal to the second addition result, taking the (i+1) th first environmental individual as the first environmental individual obtained after the sampling process;

stopping the iterative process when the j is equal to N;

8. The method according to claim 2, wherein the converting the environmental individuals in the sub-environmental set to obtain a converted sub-environmental set including a plurality of third environmental individuals, includes:

performing coding value exchange processing on at least two environment individuals in the sub-environment set to obtain a sub-environment set comprising exchanged environment individuals;

9. The method according to claim 8, wherein said performing the encoding value exchange processing on at least two environmental individuals in the sub-environmental set to obtain a sub-environmental set including the exchanged environmental individuals includes:

performing the following processing for any individual environment in the sub-environment set:

10. The method according to claim 8, wherein the performing the encoding value transformation processing on the environmental individuals in the sub-environment set including the exchanged environmental individuals, using the sub-environment set including the transformed environmental individuals as the transformed sub-environment set including the plurality of third environmental individuals, includes:

sampling the environmental individuals in the sub-environmental set comprising the exchanged environmental individuals based on the first transformation probability to obtain the environmental individuals to be transformed;

performing random transformation processing based on the coded values of the environmental individuals to be transformed in any time period based on the second variation probability to obtain transformed environmental individuals;

11. A plant growing environment control apparatus, the apparatus comprising:

a building module for performing the following process for any one of a plurality of environmental variables applied to a plant growing environment at a first time period, wherein the first time period includes a plurality of time periods; acquiring a variable value of the environment variable in any one of the time periods; performing coding processing on the variable values in the time periods to obtain coded values in the time periods; based on the sequence of the time periods, performing splicing processing on the coded values in the time periods to obtain coded data of the environment variable in the first period; splicing the coded data of the plurality of environment variables in the first period to obtain a first environment individual comprising the plurality of environment variables; copying the first environmental individuals, and taking a set of a plurality of environmental individuals obtained by the copying as a first environmental set;

12. An electronic device, the electronic device comprising:

a memory for storing executable instructions;

a processor for implementing the artificial intelligence based plant growth environment control method of any one of claims 1 to 10 when executing executable instructions stored in the memory.

13. A computer readable storage medium storing executable instructions for implementing the artificial intelligence based plant growth environment control method of any one of claims 1 to 10 when executed by a processor.