CN116302957A - Supercritical unit early warning model testing method and device based on big data platform - Google Patents

Abstract

This application relates to a supercritical unit early warning model testing method and device based on a big data platform. The specific scheme is as follows: obtain the respective measurement point data of multiple mathematical models; verify and process multiple measurement points based on the respective measurement point names of multiple measurement points based on the mathematical model; Input the value into the mathematical model to obtain the early warning result output by the mathematical model and the trigger time of the mathematical model; determine whether the operation data of the supercritical unit meets the preset trigger condition of the mathematical model at the trigger time, and obtain the first intermediate result; respond to Any one or more of the first intermediate result, trigger time and early warning result does not meet the preset requirements, and the mathematical model is adjusted for parameter adjustment. The application improves the accuracy and timeliness of the early warning of the supercritical unit through the mathematical model.

Description

Supercritical unit early warning model testing method and device based on big data platform

technical field

This application relates to the technical field of supercritical unit big data platform, in particular to a supercritical unit early warning model testing method and device based on the big data platform.

Background technique

In related technologies, thermal power plants have the characteristics of complex systems, various equipment, refined process flow and safety, so that the staff must be careful and rigorous in on-site operations and remote operations in the central control room in order to maintain the most basic functions of thermal power plants. normal operation. And when related systems and equipment fail, if the operating personnel cannot perform remote operations in the central control room to shut down the faulty equipment and open the corresponding protection equipment at the first time, it is likely to lead to further expansion, danger and seriousness of the accident. . At present, thermal power plants are dealing with the above problems by dividing the boundary conditions and boundary information of each system and equipment to carry out mechanism level judgment and early warning. This method is usually based on factory parameter values provided by equipment manufacturers and rich experience and technology. The expert experience reference value drawn up by the personnel and the reasonable value of the previous safe operation of related units of the same type are used as the basis of the mechanism analysis library to carry out different types of fault early warning judgments and divisions for each working condition, system and equipment of the thermal power plant to realize fault early warning. However, thermal power plants are composed of complex system equipment, and the above early warning methods are often based on the division of thresholds and working conditions for individual equipment, which are single-targeted and have limited adaptability.

Contents of the invention

To this end, the application provides a supercritical unit early warning model testing method and device based on a big data platform. The technical scheme of the application is as follows:

According to the first aspect of the embodiments of the present application, a supercritical unit early warning model testing method based on a big data platform is provided, the method comprising:

Obtain the respective measuring point data of a plurality of mathematical models; wherein, the plurality of mathematical models are machine learning models pre-trained based on the historical operation data of the supercritical unit; the measuring point data includes measuring point name, measuring point identification and monitoring values;

For each mathematical model, verifying the multiple measuring points based on the respective measuring point names of the multiple measuring points of the mathematical model;

In response to the verification of the plurality of measuring points, input the monitoring value into the mathematical model, and obtain the early warning result output by the mathematical model and the trigger time of the mathematical model;

Determine whether the operation data of the supercritical unit meets the preset trigger condition of the mathematical model at the trigger time, obtain a first intermediate result, and determine whether the first intermediate result meets the preset requirement;

determining whether the trigger time meets a preset requirement;

determining whether the pre-warning result meets preset requirements;

In response to any one or more of the first intermediate result, the trigger time, and the early warning result failing to meet a preset requirement, parameter adjustment processing is performed on the mathematical model.

According to an embodiment of the present application, the determining whether the operation data of the supercritical unit satisfies the preset trigger condition of the mathematical model at the trigger time to obtain a first intermediate result includes:

Determine at least one preset trigger condition of the mathematical model based on the mathematical model;

Obtain the operation data of the supercritical unit at the trigger time;

Based on the operating data, acquiring operating information respectively corresponding to the at least one preset trigger condition;

Determine whether each operation information satisfies its corresponding preset trigger condition;

In response to each piece of operation information meeting its corresponding preset trigger condition, determining that the first intermediate result is that the trigger time meets a preset requirement;

In response to at least one piece of running information failing to meet a corresponding preset trigger condition, it is determined that the first intermediate result is that the trigger time fails to meet a preset requirement.

According to an embodiment of the present application, the determining whether the trigger time meets preset requirements includes:

Acquiring a preset trigger period of the mathematical model;

Comparing the preset trigger period and the trigger time to obtain a first comparison result;

In response to the first comparison result being that the trigger time does not fall within the preset trigger period, it is determined that the trigger time does not meet a preset requirement.

According to an embodiment of the present application, said respectively determining whether said warning result meets preset requirements includes:

Acquiring the actual early warning result corresponding to the measuring point data;

Comparing the early warning result with the actual early warning result to obtain a second comparison result;

In response to the second comparison result being that the early warning result is not consistent with the actual early warning result, it is determined that the early warning result does not meet a preset requirement.

According to an embodiment of the present application, the respective measuring point names of the multiple measuring points based on the mathematical model are respectively verified for the multiple measuring points, including:

Based on the respective measuring point names and measuring point identifications of the multiple measuring points of the mathematical model, searching for the respective historical data of the multiple measuring point identifications of the mechanism model pre-stored in the database;

In response to not finding the historical data of at least one measuring point in the database, based on the measuring point name of the at least one measuring point, determining the relevant measuring point of the at least one measuring point among the plurality of measuring points;

Find the historical data of the relevant measuring points pre-stored in the database;

In response to not finding the historical data of the relevant measuring points pre-stored in the database, it is determined that the at least one measuring point has not passed the verification, performing parameter adjustment processing on the at least one measuring point respectively, and repeatedly performing the searching The step of identifying the respective historical data of the plurality of measuring points of the mechanism model pre-stored in the database.

According to an embodiment of the present application, the historical operation data of the supercritical unit includes any one or more of the following: historical operation data of steam turbines, historical operation data of boilers, historical operation data of electrical appliances, and historical operation data of thermal control 1. Chemical historical operation data; wherein, each type of historical operation data includes shutdown data, start-up process data, and operation data; the shutdown data, startup process data, and operation data all include normal data and abnormal data.

According to the second aspect of the embodiment of the present application, a supercritical unit early warning model testing device based on a big data platform is provided, the device comprising:

The obtaining module is used to obtain the respective measuring point data of a plurality of mathematical models; wherein, the plurality of mathematical models are all machine learning models pre-trained based on the historical operation data of the supercritical unit; the measuring point data includes the measuring point Name, measuring point identification and monitoring value;

A verification module, for each mathematical model, respectively verifying the multiple measuring points based on the respective measuring point names of the multiple measuring points of the mathematical model;

An input module, configured to input the monitoring value into the mathematical model in response to the verification of the plurality of measuring points, and obtain the early warning result output by the mathematical model and the trigger time of the mathematical model;

The first determination module is used to determine whether the operation data of the supercritical unit satisfies the preset trigger condition of the mathematical model at the trigger time, obtain a first intermediate result, and determine whether the first intermediate result meets the preset requirement ;

A second determination module, configured to determine whether the trigger time meets preset requirements;

The third determination module is used to determine whether the early warning result meets the preset requirements;

The parameter adjustment module is configured to perform parameter adjustment processing on the mathematical model in response to any one or more of the first intermediate result, the trigger time, and the early warning result failing to meet a preset requirement.

According to a third aspect of the embodiments of the present application, there is provided an electronic device, including: a processor, and a memory communicatively connected to the processor;

the memory stores computer-executable instructions;

The processor executes the computer-implemented instructions stored in the memory to implement the method according to any one of the first aspects.

According to a fourth aspect of the embodiments of the present application, there is provided a computer-readable storage medium, the computer-readable storage medium stores computer-executable instructions, and the computer-executable instructions are used to implement the first aspect when executed by a processor. any one of the methods described.

According to a fifth aspect of the embodiments of the present application, there is provided a computer program product, which is characterized by comprising a computer program, and when the computer program is executed by a processor, implements the method described in any one of the first aspects.

The technical solutions provided by the embodiments of the present application bring at least the following beneficial effects:

By obtaining the respective measuring point data of multiple mathematical models; for each mathematical model, verifying the multiple measuring points based on the respective measuring point names of the multiple measuring points of the mathematical model; in response to the verification of multiple measuring points , input the monitoring value into the mathematical model, obtain the early warning result output by the mathematical model and the trigger time of the mathematical model; determine whether the operation data of the supercritical unit meets the preset trigger condition of the mathematical model at the trigger time, and obtain the first intermediate result , determine whether the first intermediate result meets the preset requirement; determine whether the trigger time meets the preset requirement; determine whether the early warning result meets the preset requirement; respond to any one or more of the first intermediate result, trigger time and early warning result If the preset requirements are not met, the mathematical model is adjusted to improve the accuracy and timeliness of early warning of supercritical units in multiple professional directions through the mathematical model.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the application.

Description of drawings

The accompanying drawings here are incorporated into the specification and constitute a part of the specification, show the embodiment consistent with the application, and are used together with the specification to explain the principle of the application, and do not constitute an improper limitation of the application.

Fig. 1 is the flow chart of a kind of supercritical unit early warning model testing method based on big data platform in the embodiment of the application;

Fig. 2 is a structural block diagram of a supercritical unit early warning model testing device based on a big data platform in the embodiment of the application;

Fig. 3 is a block diagram of an electronic device in an embodiment of the present application.

Detailed ways

In order to enable ordinary persons in the art to better understand the technical solutions of the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the accompanying drawings.

It should be noted that the terms "first" and "second" in the description and claims of the present application and the above drawings are used to distinguish similar objects, but not necessarily used to describe a specific sequence or sequence. It is to be understood that the data so used are interchangeable under appropriate circumstances such that the embodiments of the application described herein can be practiced in sequences other than those illustrated or described herein. The implementations described in the following exemplary embodiments do not represent all implementations consistent with this application. Rather, they are merely examples of apparatuses and methods consistent with aspects of the present application as recited in the appended claims.

It should be noted that in related technologies, thermal power plants have the characteristics of complex systems, numerous equipment, refined process and safety, so that the staff must be careful and rigorous when operating on-site and remotely in the central control room in order to maintain firepower. The most basic normal operation of a power plant. And when related systems and equipment fail, if the operating personnel cannot perform remote operations in the central control room to shut down the faulty equipment and open the corresponding protection equipment at the first time, it is likely to lead to further expansion, danger and seriousness of the accident. . At present, thermal power plants are dealing with the above problems by dividing the boundary conditions and boundary information of each system and equipment to carry out mechanism level judgment and early warning. This method is usually based on factory parameter values provided by equipment manufacturers and rich experience and technology. The expert experience reference value drawn up by the personnel and the reasonable value of the previous safe operation of related units of the same type are used as the basis of the mechanism analysis library to carry out different types of fault early warning judgments and divisions for each working condition, system and equipment of the thermal power plant to realize fault early warning. However, thermal power plants are composed of complex system equipment, and the above early warning methods are often based on the division of thresholds and working conditions for individual equipment, which are single-targeted and have limited adaptability.

Based on the above problems, this application proposes a supercritical unit early warning model testing method and device based on a big data platform, which can realize the respective measurement point data of multiple mathematical models; for each mathematical model, based on multiple mathematical models The respective measuring point names of each measuring point are respectively verified for multiple measuring points; in response to the verification of multiple measuring points, the monitoring values are input into the mathematical model, and the early warning results output by the mathematical model and the trigger time of the mathematical model are obtained ; Determine whether the operation data of the supercritical unit meets the preset trigger condition of the mathematical model at the trigger time, obtain the first intermediate result, and determine whether the first intermediate result meets the preset requirement; determine whether the trigger time meets the preset requirement; determine the early warning Whether the result meets the preset requirement; in response to any one or more of the first intermediate result, trigger time and early warning result failing to meet the preset requirement, the mathematical model is adjusted. Therefore, the accuracy and timeliness of early warning of supercritical units in multiple professional directions through mathematical models are improved.

Fig. 1 is a flowchart of a supercritical unit early warning model testing method based on a big data platform in an embodiment of the present application.

As shown in Figure 1, the supercritical unit early warning model testing method based on the big data platform includes:

In step 101, the respective measuring point data of multiple mathematical models are obtained.

Wherein, in the embodiment of the present application, the multiple mathematical models are machine learning models pre-trained based on the historical operation data of the supercritical unit.

Wherein, in the embodiment of the present application, the measuring point data includes a measuring point name, a measuring point identifier and a monitoring value.

In some embodiments of the present application, the historical operation data of the supercritical unit includes any one or more of the following: historical operation data of steam turbines, historical operation data of boilers, historical operation data of electrical appliances, historical operation data of thermal control, chemical Class historical operation data; wherein, each type of historical operation data includes shutdown data, start-up process data, and operation data; shutdown data, startup process data, and operation data include normal data and abnormal data.

As an example of possible implementation, according to the characteristics of multiple systems in power plants, the modeling direction is divided into five categories, including: mathematical modeling of steam turbine professional direction, mathematical modeling of boiler professional direction, mathematical modeling of electrical professional direction, thermal control Mathematical modeling for professional direction, mathematical modeling for chemical major.

Extract the operating data in the system in 5 modeling directions, including: shutdown data, start-up process data, and operating data, these three categories, of which:

Shutdown data includes: unit shutdown data for short-term maintenance, unit shutdown data for long-term maintenance, unit shutdown data for short-term accident status, unit shutdown data for long-term accident status, unit short-term normal shutdown data, and unit long-term normal shutdown data;

Start-up process data include: unit cold-state start-up process data, unit warm-state start-up process data, unit hot-state start-up process data, unit extremely hot-state start-up process data, unit cold-state start-up process accident data, unit warm-state start-up process accident The data under the state, the data under the accident state during the hot start process of the unit, and the data under the accident state during the extremely hot start process of the unit;

The operating data include: normal operation data of the unit at low load (0MW-105MW), normal operation data of the unit at high load (105.1MW-297.5MW), normal operation data of the unit at full load (297.6MW-350MW), low load ( 0MW-105MW) operation data, high-load (105.1MW-297.5MW) operation data under unit accident state, and full-load (297.6MW-350MW) operation data under unit accident state.

Optionally, normal data and accident data can be found from shutdown data, startup process data, and operation data:

Normal data: unit shutdown data for short-term maintenance, unit shutdown data for long-term maintenance, unit short-term normal shutdown data, unit long-term normal shutdown data, unit cold start process data, unit warm start process data, unit hot start process data, The start-up process data of the unit in extremely hot state, the normal operation data of the unit at low load (0MW-105MW), the normal operation data of the unit at high load (105.1MW-297.5MW), and the normal operation data of the unit at full load (297.6MW-350MW);

Data in the accident state: shutdown data in the short-term accident state of the unit, shutdown data in the long-term accident state of the unit, data in the accident state during the cold start process of the unit, data in the accident state in the warm state start process of the unit, and data in the accident state in the hot state start process of the unit , The data under the accident state during the extremely hot start-up process of the unit, the low load (0MW-105MW) operation data under the accident state of the unit, the high load (105.1MW-297.5MW) operation data under the accident state of the unit, the full load (297.6MW) under the accident state of the unit MW-350MW) operating data.

Optionally, linear regression algorithm, support vector machine algorithm, nearest neighbor/k-nearest neighbor algorithm, logistic regression algorithm, decision tree algorithm, k-means algorithm, random forest algorithm, naive Bayesian algorithm, dimensionality reduction algorithm, Any one or more algorithms in the gradient enhancement algorithm are used for model training and learning based on normal data and accident state data, so that the mathematical model can learn to identify the normal operation state and accident state of the unit, and complete the establishment of the mathematical model.

Optionally, after the establishment of the mathematical model, a piece of data in the accident state can be randomly selected as a verification set to test the function of the mathematical model. The test success condition is based on the early warning issued by the model when the accident state is about to be judged. The mathematical model test is successful; if the early warning is not timely, the early warning is missed, the early warning is falsely reported, the early warning is over-reported, and the early warning is under-reported, etc., it is considered that the mathematical model test is unsuccessful.

Step 102 , for each mathematical model, verifying the multiple measuring points based on the respective measuring point names of the multiple measuring points in the mathematical model.

In some embodiments of the present application, step 102 includes:

Step a1, based on the respective measuring point names and measuring point identifiers of the multiple measuring points of the mathematical model, look up the respective historical data of the multiple measuring point identifiers of the mechanism model pre-stored in the database.

Optionally, the measuring point identifier may be the power plant identification system code of the measuring point.

Step a2, in response to not finding the historical data of at least one measuring point in the database, based on the measuring point name of the at least one measuring point, determining the relevant measuring point of at least one measuring point among the plurality of measuring points.

As an example of a possible implementation, the historical data of the measuring point is verified in the big data platform through the power plant identification system code of the measuring point of the corresponding mechanism model, and the correctness, accuracy and authenticity of the measuring point are judged. Based on the respective measuring point names and measuring point identifications of the multiple measuring points of the mechanism model, it is searched whether the historical data of the multiple measuring point identifications of the organic mechanism model are stored in the database, and in response to not finding at least one measuring point in the database Historical data, indicating that the measuring point does not meet the accuracy requirement, based on the measuring point name of at least one measuring point, determining the related measuring point of at least one measuring point among the plurality of measuring points.

Optionally, according to the name of the measuring point, the keywords, keywords and synonyms of the keywords in the name of the measuring point can be determined, and the related measuring points of the measuring point can be found based on the keywords, keywords and synonyms of the keywords. The properties of the detection value of are consistent with the properties of the detection value of the above-mentioned measuring points. It is understandable that during the data transmission process, due to network instability and other reasons, the data transmission may be incomplete or distorted. Therefore, it is necessary to verify the power plant identification system code of the measuring point, and multiple measuring points may monitor the same value. , but the names of the measuring points are slightly different, so the integrity of the data can be verified through the relevant measuring points.

In the embodiment of the present application, in response to finding the historical data of at least one measuring point in the database, it is determined that at least one measuring point passes the verification.

Step a3, searching for the historical data of the relevant measurement points pre-stored in the database.

Step a4, in response to not finding the historical data of the relevant measuring points pre-stored in the database, determining that at least one measuring point has not passed the verification, performing parameter adjustment processing on at least one measuring point, and repeatedly performing the search for the pre-stored mechanism model in the database Steps to identify the respective historical data of multiple measurement points.

As an example of a possible implementation, in response to not finding the historical data of the relevant measuring points pre-stored in the database, it is determined that at least one measuring point has not passed the verification, and at least one measuring point is adjusted respectively, and the search is repeated in the database. A step of identifying the respective historical data of a plurality of measuring points of the pre-stored mechanism model.

Step 103 , in response to the verification of multiple measuring points, input the monitoring value into the mathematical model, and obtain the early warning result output by the mathematical model and the trigger time of the mathematical model.

Step 104, determine whether the operation data of the supercritical unit meets the preset trigger condition of the mathematical model at the trigger time, obtain a first intermediate result, and determine whether the first intermediate result meets the preset requirement.

In some embodiments of the present application, step 104 includes:

Step b1, based on the mathematical model, determine at least one preset trigger condition of the mathematical model.

It is understandable that since multiple mathematical models belong to different professional directions of the power plant, it is necessary to formulate corresponding model trigger rules, so that the models can be used more flexibly, more quickly, and more conveniently.

For example, for the coal mill body blockage model of the pulverization system and the coal pulverizer outlet pulverizer pipe blockage model of the pulverization system, the model trigger rules can be coal mill operation, coal feeder operation, and coal supply normal; lubrication For the fuel tank leakage model of the oil system and the fuel tank leakage model of the high-pressure anti-fuel system, the triggering rules of the models can be as follows: The liquid level of the high-pressure heater is normal, the pressure of the main pipe, and the flow rate are normal; the first-stage reverse osmosis fouling model of the reverse osmosis system, and the second-stage reverse osmosis fouling model of the reverse osmosis system, the model triggering rules can be the total water production, the first-stage reverse osmosis The pressure difference and the differential pressure of the second-stage reverse osmosis are normal.

Step b2, obtaining the operation data of the supercritical unit at the trigger time.

Step b3, based on the operating data, acquiring operating information corresponding to at least one preset trigger condition.

Step b4, determining whether each piece of operation information satisfies its corresponding preset trigger condition.

It can be understood that the operation data of the supercritical unit at the trigger time is obtained, the operation information corresponding to at least one preset trigger condition is obtained based on the operation data, and whether the current operation state of the supercritical unit satisfies the triggering of the mathematical model is determined based on the operation information. condition.

In step b5, in response to each piece of operating information meeting its corresponding preset trigger condition, it is determined that the first intermediate result is that the trigger time meets the preset requirement.

Step b6, in response to at least one piece of operating information not meeting the respective corresponding preset trigger conditions, determining that the first intermediate result is that the trigger time does not meet the preset requirements.

Step 105, determine whether the trigger time meets the preset requirement.

In some embodiments of the present application, step 105 includes:

Step c1, obtaining the preset trigger period of the mathematical model.

Step c2, comparing the preset trigger period with the trigger time to obtain a first comparison result.

Step c3, in response to the first comparison result that the trigger time does not fall within the preset trigger period, it is determined that the trigger time does not meet the preset requirement.

As an example of a possible implementation, since each mathematical model is different for different accident judgment periods, the trigger periods and trigger rules can be divided into different categories.

For example, the preset trigger period of the mathematical model may include:

Short-term quick trigger accident warning, set the time period within 10 seconds;

The short-term gradual rise triggers the accident warning, and the set time period is within 10 seconds to 10 minutes;

The medium-term uniform speed change triggers an accident warning, and the set time period is within 10 minutes to 60 minutes;

The long-term slow change triggers an accident warning, and the set time period is within 60 minutes to 600 minutes.

For example, the blockage of the coal mill body in the pulverizing system and the clogging of the powder pipe at the outlet of the coal pulverizer in the pulverizing system are short-term rapid trigger accident warnings. The accidents occur very quickly, and a short period of time needs to be set in the model judgment cycle;

For example, the fuel tank leakage of the lubricating oil system and the fuel tank leakage of the high-pressure anti-fuel system are all short-term and short-term gradual rises that trigger accident warnings. The accident situation will gradually rise in a short period of time, and a short period of time needs to be set in the model judgment cycle;

For example, the leakage of the high-pressure heater in the high-pressure system belongs to the medium-term uniform-speed change that triggers the accident warning, and the accident situation will gradually increase within a period of time, and a period of time needs to be set within the model judgment period;

For example, reverse osmosis fouling in the first stage of the reverse osmosis system and reverse osmosis fouling in the second stage of the reverse osmosis system are both long-term slow changes that trigger accident warnings. The accident situation will gradually increase over a long period of time, and it is necessary to set a period in the model judgment cycle. long time.

Step 106, determine whether the warning result meets the preset requirements.

In some embodiments of the present application, step 106 includes:

Step d1, obtaining the actual early warning results corresponding to the measuring point data.

Step d2, comparing the warning result with the actual warning result to obtain a second comparison result.

Step d3, in response to the second comparison result being that the warning result is not consistent with the actual warning result, it is determined that the warning result does not meet the preset requirement.

Step 107, in response to any one or more of the first intermediate result, the trigger time and the early warning result failing to meet the preset requirements, perform parameter adjustment processing on the mathematical model.

For example, taking the leak warning of the high-pressure anti-fuel system specialized in steam turbines as an example, the oil level of the high-pressure anti-fuel tank, the pressure of the main pipe, and the operation data of the oil pump are used as training data, and the training data is divided into shutdown data, start-up process data, and operation data. Data, the data of these three stages are used as the training set data and the verification set data, and the convolutional neural network algorithm is used to build the model, and finally the model is verified. Next, deploy the operator block of the high-pressure anti-fuel system leakage early warning model to the big data platform. It is necessary to check the input measurement points on the platform, including: the oil level of the high-pressure anti-fuel oil tank, the pressure of the main pipe, and the operation of the oil pump, etc., and set the judgment cycle Within 10 seconds to 10 minutes, set the trigger rule of the model as the oil pump is running, the oil level of the fuel tank is normal, and the pressure of the main oil pipe is normal. Finally, the fault data is used to test and verify the operator block of the mathematical model.

According to the supercritical unit early warning model testing method based on the big data platform of the embodiment of the present application, by obtaining the respective measuring point data of a plurality of mathematical models; for each mathematical model, the respective measuring point names of a plurality of measuring points based on the mathematical model Perform verification processing on multiple measuring points respectively; in response to the verification of multiple measuring points, input the monitoring value into the mathematical model, obtain the early warning result output by the mathematical model and the trigger time of the mathematical model; determine the supercritical condition at the trigger time Whether the operating data of the unit meets the preset trigger conditions of the mathematical model, obtain the first intermediate result, and determine whether the first intermediate result meets the preset requirements; determine whether the trigger time meets the preset requirements; determine whether the early warning results meet the preset requirements; respond When any one or more of the first intermediate result, the trigger time and the early warning result do not meet the preset requirements, the mathematical model is adjusted. Therefore, the accuracy and timeliness of early warning of supercritical units in multiple professional directions through mathematical models are improved.

FIG. 2 is a flow chart of a supercritical unit early warning model testing device based on a big data platform in an embodiment of the present application.

As shown in Figure 2, the supercritical unit early warning model test device based on the big data platform includes:

Acquisition module 201 is used to obtain the respective measuring point data of a plurality of mathematical models; wherein, a plurality of mathematical models are machine learning models pre-trained based on historical operation data of supercritical units; measuring point data includes measuring point name, measuring point Point identification and monitoring values;

The verification module 202 is used for verifying multiple measuring points based on the respective measuring point names of the multiple measuring points of the mathematical model for each mathematical model;

The input module 203 is used to input the monitoring value into the mathematical model in response to the verification of multiple measuring points, and obtain the early warning result output by the mathematical model and the trigger time of the mathematical model;

The first determination module 204 is used to determine whether the operation data of the supercritical unit meets the preset trigger condition of the mathematical model at the trigger time, obtain the first intermediate result, and determine whether the first intermediate result meets the preset requirement;

The second determination module 205 is used to determine whether the trigger time meets the preset requirements;

The third determination module 206 is used to determine whether the early warning result meets the preset requirements;

The parameter adjustment module 207 is configured to perform parameter adjustment processing on the mathematical model in response to any one or more of the first intermediate result, the trigger time and the early warning result failing to meet the preset requirements.

According to the supercritical unit early warning model testing device based on the big data platform of the embodiment of the application, by obtaining the respective measuring point data of a plurality of mathematical models; for each mathematical model, the respective measuring point names of a plurality of measuring points based on the mathematical model Perform verification processing on multiple measuring points respectively; in response to the verification of multiple measuring points, input the monitoring value into the mathematical model, obtain the early warning result output by the mathematical model and the trigger time of the mathematical model; determine the supercritical condition at the trigger time Whether the operating data of the unit meets the preset trigger conditions of the mathematical model, obtain the first intermediate result, and determine whether the first intermediate result meets the preset requirements; determine whether the trigger time meets the preset requirements; determine whether the early warning results meet the preset requirements; respond When any one or more of the first intermediate result, the trigger time and the early warning result do not meet the preset requirements, the mathematical model is adjusted. Therefore, the accuracy and timeliness of early warning of supercritical units in multiple professional directions through mathematical models are improved.

Fig. 3 is a block diagram of an electronic device in an embodiment of the present application. As shown in FIG. 3 , the electronic device may include: a transceiver 31 , a processor 32 , and a memory 33 .

The processor 32 executes the computer-executable instructions stored in the memory, so that the processor 32 executes the solutions in the above-mentioned embodiments. Processor 32 can be a general-purpose processor, including a central processing unit CPU, a network processor (network processor, NP) etc.; it can also be a digital signal processor DSP, an application specific integrated circuit ASIC, a field programmable gate array FPGA or other programmable Logic devices, discrete gate or transistor logic devices, discrete hardware components.

The memory 33 is connected to the processor 32 through the system bus and communicates with each other, and the memory 33 is used for storing computer program instructions.

The transceiver 31 can be used to acquire tasks to be executed and configuration information of the tasks to be executed.

The system bus may be a peripheral component interconnect standard (peripheral component interconnect, PCI) bus or an extended industry standard architecture (extended industry standard architecture, EISA) bus or the like. The system bus can be divided into address bus, data bus, control bus and so on. For ease of representation, only one thick line is used in the figure, but it does not mean that there is only one bus or one type of bus. Transceivers are used to enable communication between database access devices and other computers such as clients, read-write libraries, and read-only libraries. The memory may include random access memory (random access memory, RAM), and may also include non-volatile memory (non-volatile memory).

The electronic device provided in the embodiment of the present application may be the terminal device in the foregoing embodiment.

The embodiment of the present application also provides a chip for running instructions, and the chip is used to implement the technical solution of the message processing method in the above embodiment.

The embodiment of the present application also provides a computer-readable storage medium, where computer instructions are stored in the computer-readable storage medium, and when the computer instructions are run on the computer, the computer is made to execute the technical solution of the message processing method of the above-mentioned embodiment.

The embodiment of the present application also provides a computer program product, the computer program product includes a computer program, which is stored in a computer-readable storage medium, at least one processor can read the computer program from the computer-readable storage medium, at least one processor The technical solutions of the message processing method in the foregoing embodiments can be realized when the computer program is executed.

Other embodiments of the present application will be readily apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This application is intended to cover any modification, use or adaptation of the application, these modifications, uses or adaptations follow the general principles of the application and include common knowledge or conventional technical means in the technical field not disclosed in the application . The specification and examples are to be considered exemplary only, with a true scope and spirit of the application indicated by the following claims.

It should be understood that the present application is not limited to the precise constructions which have been described above and shown in the accompanying drawings, and various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims (10)

1. a supercritical unit early warning model testing method based on big data platform, is characterized in that, described method comprises: Obtain the respective measuring point data of a plurality of mathematical models; wherein, the plurality of mathematical models are machine learning models pre-trained based on the historical operation data of the supercritical unit; the measuring point data includes measuring point name, measuring point identification and monitoring values; For each mathematical model, verifying the multiple measuring points based on the respective measuring point names of the multiple measuring points of the mathematical model; In response to the verification of the plurality of measuring points, input the monitoring value into the mathematical model, and obtain the early warning result output by the mathematical model and the trigger time of the mathematical model; Determine whether the operation data of the supercritical unit meets the preset trigger condition of the mathematical model at the trigger time, obtain a first intermediate result, and determine whether the first intermediate result meets the preset requirement; determining whether the trigger time meets a preset requirement; determining whether the pre-warning result meets preset requirements; In response to any one or more of the first intermediate result, the trigger time, and the early warning result failing to meet a preset requirement, parameter adjustment processing is performed on the mathematical model. 2. The method according to claim 1, wherein said determining whether the operating data of the supercritical unit satisfies the preset trigger condition of the mathematical model at the trigger time, obtains the first intermediate result, comprising: Determine at least one preset trigger condition of the mathematical model based on the mathematical model; Obtain the operation data of the supercritical unit at the trigger time; Based on the operating data, acquiring operating information respectively corresponding to the at least one preset trigger condition; Determine whether each operation information satisfies its corresponding preset trigger condition; In response to each piece of operation information meeting its corresponding preset trigger condition, determining that the first intermediate result is that the trigger time meets a preset requirement; In response to at least one piece of running information failing to meet a corresponding preset trigger condition, it is determined that the first intermediate result is that the trigger time fails to meet a preset requirement. 3. The method according to claim 1, wherein the determining whether the trigger time meets a preset requirement comprises: Acquiring a preset trigger period of the mathematical model; Comparing the preset trigger period and the trigger time to obtain a first comparison result; In response to the first comparison result being that the trigger time does not fall within the preset trigger period, it is determined that the trigger time does not meet a preset requirement. 4. The method according to claim 1, wherein said respectively determining whether said early warning results meet preset requirements comprises: Acquiring the actual early warning result corresponding to the measuring point data; Comparing the early warning result with the actual early warning result to obtain a second comparison result; In response to the second comparison result being that the early warning result is not consistent with the actual early warning result, it is determined that the early warning result does not meet a preset requirement. 5. The method according to claim 1, characterized in that, the respective measuring point titles of the multiple measuring points based on the mathematical model are respectively verified for the plurality of measuring points, including: Based on the respective measuring point names and measuring point identifications of the multiple measuring points of the mathematical model, searching for the respective historical data of the multiple measuring point identifications of the mechanism model pre-stored in the database; In response to not finding the historical data of at least one measuring point in the database, based on the measuring point name of the at least one measuring point, determining the relevant measuring point of the at least one measuring point among the plurality of measuring points; Find the historical data of the relevant measuring points pre-stored in the database; In response to not finding the historical data of the relevant measuring points pre-stored in the database, it is determined that the at least one measuring point has not passed the verification, performing parameter adjustment processing on the at least one measuring point respectively, and repeatedly performing the searching The step of identifying the respective historical data of the plurality of measuring points of the mechanism model pre-stored in the database. 6. The method according to claim 1, wherein the historical operating data of the supercritical unit includes any one or more of the following: historical operating data of steam turbines, historical operating data of boilers, historical operating data of electrical appliances , thermal control historical operation data, and chemical historical operation data; wherein, each type of historical operation data includes shutdown data, startup process data, and operation data; the shutdown data, startup process data, and operation data include normal data and abnormal data. 7. A supercritical unit early warning model testing device based on a big data platform, characterized in that the device comprises: The obtaining module is used to obtain the respective measuring point data of a plurality of mathematical models; wherein, the plurality of mathematical models are all machine learning models pre-trained based on the historical operation data of the supercritical unit; the measuring point data includes the measuring point Name, measuring point identification and monitoring value; A verification module, for each mathematical model, respectively verifying the multiple measuring points based on the respective measuring point names of the multiple measuring points of the mathematical model; An input module, configured to input the monitoring value into the mathematical model in response to the verification of the plurality of measuring points, and obtain the early warning result output by the mathematical model and the trigger time of the mathematical model; The first determination module is used to determine whether the operation data of the supercritical unit satisfies the preset trigger condition of the mathematical model at the trigger time, obtain a first intermediate result, and determine whether the first intermediate result meets the preset requirement ; A second determination module, configured to determine whether the trigger time meets preset requirements; The third determination module is used to determine whether the early warning result meets the preset requirements; The parameter adjustment module is configured to perform parameter adjustment processing on the mathematical model in response to any one or more of the first intermediate result, the trigger time, and the early warning result failing to meet a preset requirement. 8. An electronic device, comprising: a processor, and a memory communicatively connected to the processor; the memory stores computer-executable instructions; The processor executes the computer-implemented instructions stored in the memory to implement the method according to any one of claims 1-6. 9. A computer-readable storage medium, characterized in that, computer-executable instructions are stored in the computer-readable storage medium, and the computer-executable instructions are used to implement any one of claims 1-6 when executed by a processor. method described in the item. 10. A computer program product, characterized by comprising a computer program, and implementing the method according to any one of claims 1-6 when the computer program is executed by a processor.

Images