JPH0340002A - Greenhouse culture control supporting device - Google Patents

Abstract

PURPOSE:To configure a system not depending on the observation capability of a cultivator by introducing a crop model function and a state sensor function. CONSTITUTION:A controller means 2 outputs the manipulated variable of a greenhouse based on a greenhouse environmental measuring value PV and a targeted environmental value SV, and the crop model function 8 outputs a virtual crop state value by inputting the manipulated variable. An expert system 5 calculates the targeted environmental value to the controller means 2 with the support of a knowledge data base by inputting the output of the crop state value of the crop model function 8. Furthermore, the expert system 5 confirms the difference of the virtual crop state value from the crop model function 8 and a real crop state to the cultivator via an interface 11, and corrects the parameter of the crop model function 8 when the difference over an allowable value exists. A state sensor means 12 supplies the state measuring value of a crop to the crop model function 8 and an expert system function. In such a way, automatic greenhouse environmental control can be performed with the crop state value by the crop model function with scarcely receiving the assistance of the cultivator.

Description

DETAILED DESCRIPTION OF THE INVENTION <Industrial Application Field> The present invention relates to an improvement of a greenhouse cultivation control support device using an Expert 1 system.

<Prior Art> An outline of a greenhouse control device using a computer will be explained based on FIG. 2.

1 is a greenhouse in which environmental conditions such as temperature, humidity, culture solution concentration, and water supply amount are managed and controlled; 2 is a controller means;
It is connected to the greenhouse 1 by a pig communication means 3 for exchanging measurement information and control information.

The controller 1 roller means 2 includes a control computer 201,
This is realized by a monitoring computer 202 and a data communication means 203 that connects them.

4 is an online data item handled within the controller means 2, which includes measurement data of various sensors, calculations based on the measurement data, integrated data, and combination data.
2. -The precepts have been broken in the evening state data, etc.

In the case where there is no support means, the controller means 2 compares the environmental condition parameter setting values set by the operator via the monitoring computer with the measured online data, and outputs the operation output so that the deviation becomes zero. The information is supplied to the operating means within the system to perform automatic control of each parameter, and further perform various manual operations.

With the controller means having such a configuration, the operator can periodically or at any timing depending on the growth status of crops in the greenhouse, set parameters that are deemed appropriate, and perform manual operations in response to changes in the situation. However, the current criteria are determined by the intuition or knack of experienced growers.

With current control systems, it is impossible to convert these intuitions and tricks into algorithms, so it has not yet been possible to completely automate crop growth management.

Therefore, the applicant, by Utility Application No. 6.2-48751,
We have provided a control device that allows an operator to accurately determine parameter settings and treatments using an expert method without relying on the experience or intuition of an expert.

Its feature is that when setting control parameters periodically or at any timing, with the support of expert capabilities, input data based on the operator's judgment, online measurement data of the greenhouse environment at the present time, and control performance over a certain period of the past can be used. Based on the data, the optimal parameter settings or treatment methods for the current greenhouse environmental conditions are inferred and output for display.

An explanation of the characteristic parts will be added based on FIG.

Reference numeral 5 denotes support means using an expert system, which consists of a combination device 501 equipped with an inference function and a data file 502 in which a knowledge base necessary for inference is constructed.

A data communication means 6 connects the controller means 2 and the support means 5, and includes a function of transmitting online data 4 necessary for inference to the support means 5 side.

Next, the operation procedure of the support function will be explained with reference to the flowchart shown in FIG.

(1) When the system is started, a data communication program is first started, takes in online data from the controller means 2, executes data processing, creates a data file, and sends it to the cause estimation expert system.

(2) At the same time, graph the current environmental condition of the greenhouse and important management items, such as one week's worth of solution supply data.
Display on RT.

(3) The cause estimation expert system accepts input of answers to questions asked by the operator and performs inference based on the knowledge base.

(4) Furthermore, the cause estimation expert system infers the current problem based on the online data and displays the cause inference results, and also converts the input answer data and online data into a file and passes it to the treatment method estimation expert system.

(5) The expert system for estimating a treatment method uses the data to infer a treatment method and displays the results on a CRT.

<Problems to be solved by the invention> With support means using such an expert system,
Since it is an offline system that draws conclusions from the information provided and issues instructions to the grower, the grower must observe and input the state of the crop.

Therefore, no matter how good the expert system is, there is a problem that if the crop condition is poorly understood, cultivation will not go well.

Furthermore, even if a judgment is made by comparing it with past data held in a database, there is a problem in that the judgment result is influenced by the quality of that data.

Furthermore, since the support provided by the Expert 1 system is of an offline configuration and is not included in the control loop, there is a problem in that automatic greenhouse environment control cannot be realized by the support means.

An object of the present invention is to provide a greenhouse cultivation control support device that eliminates such problems.

Means for Solving Problems> The first structural feature of the present invention is that in a crop cultivation control device in a greenhouse, a controller means outputs a manipulated variable of the greenhouse environment based on a greenhouse environment measurement value and a target environment value; ,
A crop model function that inputs the above operation amount and the crop state value force of this model function calculates the above target environmental value with the support of a knowledge database, and also calculates the difference between the above crop state value and the actual crop state. The present invention also includes an expert system function that has the function of checking the parameters of the crop model function with the grower and correcting the parameters of the crop model function if there is a difference that exceeds an allowable value.

A second feature is that, in addition to the above configuration, the present invention includes a state sensor means for supplying measured values of the state of the crop to the crop model function and the expert system function.

Effect> The controller means outputs a manipulated variable of the greenhouse environment based on the greenhouse environment measurement value and the target environment value.

The crop model function inputs a manipulated variable and outputs a virtual crop state value.

The expert system inputs the crop state values and values of the crop model function and calculates the target environmental values to the controller means with the aid of the knowledge database.

Additionally, the expert system confirms with the grower via an interface the difference between the virtual crop condition values from the crop model function and the actual crop condition, and if the difference is more than acceptable, the expert system changes the parameters of the crop model function. Correct.

The condition sensor means provides crop condition measurements to the crop model function and the expert system function.

<Example> The present invention will be explained in detail based on FIG.

Elements explained in FIG. 2 are shown with the same reference numerals.

1 is a greenhouse in which environmental conditions such as temperature, humidity, CO2 concentration, and soil moisture are controlled.

2 is a controller means, and the target environmental value SV is determined by data communication from the expert system 5, which is a support means.
and the deviation E of the environmental measurement value PV of the greenhouse are controlled and calculated, and the greenhouse environment is controlled by data communication using the operation engine MV.

Reference numeral 4 denotes an online data item handled by the controller means, which is composed of measurement data of various sensors, calculation data based on the measurement data, and computer status data.

Such a basic loop! The I4 precepts are the same as in the prior art.

7 is a crop grown in a greenhouse, and 8 is a crop model function that simulates various status items of this crop.It inputs online data of the greenhouse and calculates current crop status management items, such as photosynthetic dry matter amount in tons. is supplied as virtual crop status value data.

An example of a basic model regarding the maximum amount of photosynthesis P and o is shown in the equation below.

Ppot-PIIlax(I/(P□ax/E1+I)
)((cCo)/(1/E o+(C1-Co)))
f (T,) where: Ppot' Estimated maximum value of photosynthesis gm-2s-1
P: Photosynthesis rate under optimal environment gm-2aX ■=Light intensity Wm-2E・: Light usage efficiency Jg-1C0: CO2 in leaves
Concentration gm-3 Peng C: CO2 compensation value gm-3-1 E: CO2 utilization efficiency mgf ('r,)
: Leaf temperature Based on the maximum photosynthetic amount from this equation, it is possible to determine the photosynthetic rate that has a predetermined numerical relationship.

Parameters that can be tuned by initial Sera) or modification commands are P, E,, CC, EC111aX
1, etc., and it is possible to tune the constant terms in the formula for calculating other parameters as well.

The parameters of the simulation model can be maintained by the man-machine means 9, and can be automatically corrected from the expert 1 to system measurements.

The expert system 5, which functions as a model for the veteran grower, communicates with the data file 502 that has accumulated the knowledge of the experts 1 to 502, and determines the optimum parameters for handling the online data items based on the crop condition value F from the crop model function 8. The environmental target value S■ is inferred and output.

Inference methods include simple IF...THEN...
Advanced inference that performs optimal evaluation using fuzzy theory that takes ambiguity into account is possible at the current technological level.

10 is a grower who actually observes the size, color, shape, etc. of the crop 7;

The expert 1-system 5 checks with the grower via the interface means 11 with man-machine means whether the fictitious crop condition value data from the crop model function 8 is consistent with the real crop condition.

In response to this confirmation instruction, the grower replies to the expert system with the observation results via the interface.

For example, an expert 1 or the system can ask the grower the environmental target value, saying, ``What is the current photosynthetic rate? Do you want to modify the target value to increase (or decrease) the rate?'' is asked before outputting, and if there is no response from the grower within 10 minutes, the information is output.

In the expert system, if there is a difference between the response content and the crop status value of the crop model function that exceeds an allowable value, model parameter correction is performed to improve the accuracy of the model.

1 In this manner, the expert system of the present invention has a learning function that constantly verifies actual crop condition data with the grower and corrects the crop model function so that it behaves the same as the actual crop.

Therefore, even if the crop model design (parameter set) accuracy is not sufficient, the accuracy of the model can be improved by the learning function after control starts, and if the accuracy improves and control becomes stable, cultivation with little intervention,
Automatic greenhouse environment control is possible using crop condition values based on the crop model function.

The route indicated by G transmits the measured values of the crop 7 by the condition sensor 12 to the crop model function 8 and the expert system 5.
This is the route that supplies the area.

This measurement value includes stem diameter, measurement data of transpiration flow, size,
This is image analysis data such as shape, etc., and by using this data, it is possible to reinforce the grower's confirmation information, making it possible to further improve the accuracy of the model, and creating a greenhouse environment that further reduces the burden on the grower. Driving is possible.

2 Examples of sensors currently available for measuring the stem system include direct measurement gauges and image analysis methods;
These include measuring the flow of water in a stem using MR, measuring photosynthesis from the amount of chlorophyll in leaves by shining light on it, and measuring the flow of water by measuring the capacitance between electrodes attached to both sides of the stem.

<Effects of the Invention> As explained above, in the control by the conventional support device,
Because the grower observes the condition of the crop and inputs the condition value into the expert system to obtain the environmental target value,
According to the present invention, by introducing a crop model function and a state sensor function, it is possible to construct a system that does not rely on the grower's observation ability.

This eliminates the problem of differences in cultivation results even when using the same system depending on the level of observation ability of the grower, and makes it possible to realize an ideal greenhouse environment control device that greatly reduces the burden on growers. You can get close.

[Brief explanation of drawings]

FIG. 1 is a block diagram showing an embodiment of the present invention, FIG. 2 is a block diagram of a greenhouse environment control device using a conventional support device, and FIG. 3 is a flow chart 1 to explain the operation thereof.

Claims (2)

[Claims] (1) In a greenhouse crop cultivation control device, a controller unit outputs a manipulated variable of the greenhouse environment based on a greenhouse environment measurement value and a target environmental value, a crop model function inputs the manipulated variable, and a crop of this model function. Input the state value output and calculate the above target environmental value with the support of the knowledge database. At the same time, confirm the difference between the above crop state value and the actual crop state with the grower, and if the difference exceeds the allowable value, the above target environmental value will be calculated. A greenhouse cultivation control support device equipped with an expert system function having a function of correcting parameters of a crop model function. (2) In a greenhouse crop cultivation control device, a controller unit outputs a manipulated variable of the greenhouse environment based on a greenhouse environment measurement value and a target environmental value, a crop model function inputs the manipulated variable, and a crop of this model function. Input the state value output and calculate the above target environmental value with the support of the knowledge database. At the same time, confirm the difference between the above crop state value and the actual crop state with the grower, and if the difference exceeds the allowable value, the above target environmental value will be calculated. A greenhouse cultivation control support device comprising: an expert system function having a function of modifying parameters of a crop model function; and a state sensor means for supplying measured values of the state of the crop to the crop model function and the expert system function.