Control method and system for fused salt energy storage system

Abstract

The invention relates to the technical field of system control, in particular to a control method and system for a fused salt energy storage system, and the method mainly comprises the steps: judging whether a first change rate is greater than zero or not, if yes, obtaining a first limit rate of each piece of first sub-data according to the first sub-data and a corresponding limit value, acquiring a second limit rate of the first sub-data of the pipeline or the heating equipment adjacent to the first sub-data of the target, establishing a sub-data evaluation model, outputting an evaluation reference value through the sub-data evaluation model, and if the evaluation reference value is higher, representing that the current target is more likely to have a risk. According to the method, the pipeline or the heating equipment corresponding to the current highest evaluation reference value is obtained, the alarm signal is sent to the remote control end, and the personnel of the remote control end decides whether to continuously and normally adjust the control valve or the heating equipment according to the predicted power demand or not. Therefore, the method can identify potential risks to a certain extent.

Description

Technical Field

[0001] The present invention relates to the technical field of system control, and more particularly, to a control method and system for a molten salt energy storage system. Background Art

[0002] A molten salt energy storage system is a technology that uses molten salt as an energy storage medium, mainly for concentrated solar power generation and the storage and utilization of other renewable energy sources. The system stores thermal energy by heating molten salt to a high temperature (usually between 250°C and 600°C). Then, this thermal energy can be converted into steam through a heat exchanger, and further drive a turbine generator to generate electricity. This includes the selection of molten salt: common molten salt combinations include sodium nitrate and potassium nitrate. Due to their high thermal conductivity and heat capacity, they become ideal energy storage media. The molten salt energy storage system can store thermal energy in a highly concentrated manner, and has a higher energy density compared to traditional water energy storage.

[0003] However, the current molten salt energy storage system does not have a good function of predicting potential problems. When a failure occurs in the molten salt energy storage system, a relatively large accident may occur, causing unnecessary losses. Therefore, there is an urgent need for a method that can predict potential risks to a certain extent and control the molten salt energy storage system. Summary of the Invention

[0004] The purpose of the present invention is to provide a control method and system for a molten salt energy storage system to solve the above problems in the prior art.

[0005] The present invention is achieved through the following technical solutions: In a first aspect, the present invention provides a control method for a molten salt energy storage system, including: Obtain historical power demand data of the energy consumption end within a target period, preprocess the historical power demand data of the energy consumption end, establish a power prediction model, and output a power demand prediction result within a target period through the preprocessed historical power demand data of the energy consumption end and the power prediction model; Set an adjustment period, divide the target period into several adjustment time periods through the adjustment period, obtain the first power demand of the target adjustment time period, and obtain the predicted second power demand of the next judgment adjustment time period adjacent to the target adjustment time period through the power demand prediction result, and obtain a first change rate of the power demand through the first power demand and the second power demand; Establish a simulation model, and simulate and obtain first sub-data of a plurality of pipeline detection data and a plurality of heating equipment data through the simulation model according to the second power demand, and respectively obtain limit values corresponding to the plurality of first sub-data; Determine whether there is a second sub-data greater than or equal to the limit value among several first sub-data. If so, obtain the pipeline or heating equipment corresponding to the second sub-data, send a control signal to maintain the current state to the control valve or heating equipment on the pipeline corresponding to the second sub-data, and at the same time send an alarm signal to the remote control terminal; If not, determine whether the first change rate is greater than zero. If it is greater than zero, obtain the first limit rate of each first sub-data according to the first sub-data and the corresponding limit value, obtain the second limit rate of the first sub-data of the pipeline or heating equipment adjacent to the target first sub-data, establish a sub-data evaluation model, and output an evaluation reference value through the sub-data evaluation model according to the first limit rate, the second limit rate, and the first change rate; Determine whether the target corresponding to the first sub-data corresponding to the maximum evaluation reference value is a heating equipment or a pipeline. If it is a pipeline, send a control signal to keep the control valve of the target pipeline in the current state and increase the heating power of the heating equipment until the output power of the molten salt energy storage system reaches the second power demand. If it is a heating equipment, send a control signal to keep the heating equipment in the current state and an alarm signal; If the first change rate is not greater than zero, normally output the control signal of the control valve, pump or heating equipment corresponding to reaching the second power demand.

[0006] Preferably, the establishment of the power prediction model includes: Record the historical power demand data of the energy consumption end as the load power sequence, fit the historical power demand data in the load power sequence to obtain the fitting function of the load growth rate; Correct the growth cycle of the electricity load through the residual value of the electricity dependence to obtain the periodic characteristics of the electricity load growth; Define the LSTM memory unit in the model. The LSTM memory unit includes a forgetting gate, an input gate and an output gate. Construct three influencing factors based on the basic information of the electricity load, input the three influencing factors into the LSTM memory unit through the input gate, and output the memory result through the analysis and memory of the LSTM memory unit at the output gate; Add the Attention mechanism to construct a target model based on the Attention mechanism and the LSTM memory unit. Analyze the probability feature vector of the attention distribution of the output value of the LSTM memory unit in the model through the Attention mechanism, and obtain the influence result of the influencing factor on the electricity load through the target model; Construct a linear regression equation of the electricity load influencing factor characteristics in the region, construct a scaling matrix for predicting the growth of the electricity load according to the linear regression equation, and perform scaling correction on the scaling matrix. On the corrected result, solve the predicted growth value of the electricity load.

[0007] Preferably, correcting the growth cycle of the electricity consumption load by the electricity-dependence residual value, the obtained periodic characteristics of the electricity consumption load growth include:

[0008]

[0009] In the formula, is the electricity-dependence residual value, is the type of the growth trend fitting function, is the function value of the growth trend fitting function, is the electricity-dependence coefficient, is the mean value of the function value within the prediction period, is the total number of types of the growth trend fitting function, is the growth cycle of the electricity consumption load, is the growth cycle, is the determination coefficient, is the fitting function.

[0010] Preferably, the three influencing factors include:

[0011]

[0012]

[0013] In the formula, is the first influencing factor of the electricity consumption load, is the second influencing factor of the electricity consumption load, is the third influencing factor of the electricity consumption load, is the real-time temperature of the current season, is the mean temperature of the current season, is the seasonal fluctuation coefficient, is the number of seasons, is the seasonal parameter, is the power supply load margin, is the deviation cancellation coefficient, is the maximum electricity consumption load difference, is the relative residual error, is the daily average load, is the valley electricity coefficient, is the daily maximum peak-valley difference, is the daily load fluctuation coefficient.

[0014] Preferably, the solving of the predicted growth value of the electricity consumption load includes:

[0015] In the formula, is the predicted growth value of the electricity load, is the power supply capacity, is the scaling matrix, is the area of the region of the load to be predicted, is the historical peak load value, is the time series parameter, is the time window length.

[0016] Preferably, the establishment of the simulation model includes: Obtain a number of historical detection data of the pipeline and the heating equipment with the increase of the power demand at the energy consumption end respectively, divide the detection data of the same type into the same data group, and obtain the relationship between the detection data and the power at the energy consumption end according to the data group; Obtain the current power demand at the energy consumption end, and output the first sub-data predicted corresponding to the current power demand at the energy consumption end through the data group.

[0017] Preferably, sending the control signal includes: Obtain the system load rate of the current molten salt energy storage system, set the system load alarm threshold, and judge whether the system load rate at the current moment reaches the system load alarm threshold; If it does not reach, send the control signal normally. If it reaches, obtain the information content to be sent by the control signal. If the control signal is the control information to maintain the current state, obtain the maximum evaluation reference value, then establish a system load rate regulation model, output the system load rate that needs to be reduced through the system load rate regulation model based on the maximum evaluation reference value, and send the control signal. If the control signal is a normal output, send the control signal normally.

[0018] Preferably, the system load rate regulation model includes:

[0019] In the formula, is the system load rate that needs to be reduced, is the maximum evaluation reference value, is the number of all evaluation reference values, is the th evaluation reference value.

[0020] Preferably, the sub-data evaluation model includes:

[0021]

[0022]

[0023] In the formula, is the evaluation reference value, is the first change rate, is the first limit rate, is the second limit rate, , , …, are respectively the first limit rates from the first to the th first limit rates of the first sub - data of the pipeline or heating equipment adjacent to the target first sub - data, is the value of the first sub - data, is the limit value corresponding to the first sub - data.

[0024] In a second aspect, the present invention also provides a control system for a molten salt energy storage system, including: A power prediction module configured to output a power demand prediction result within a target period through pre - processed historical power demand data of the energy - consuming end and a power prediction model; A demand change module configured to obtain a first change rate of the power demand through a first power demand and a second power demand; A system simulation module configured to establish a simulation model and simulate to obtain first sub - data of several pipeline detection data and several heating equipment data; A first judgment module configured to judge whether there is a second sub - data greater than or equal to the limit value among several first sub - data. If so, obtain the pipeline or heating equipment corresponding to the second sub - data, send a control signal to keep the current state to the control valve or heating equipment on the pipeline corresponding to the second sub - data, and simultaneously send an alarm signal to the remote control terminal; A second judgment module configured to, if not, judge whether the first change rate is greater than zero. If it is greater than zero, obtain the first limit rate of each first sub - data according to the first sub - data and the corresponding limit value, obtain the second limit rate of the first sub - data of the pipeline or heating equipment adjacent to the target first sub - data, establish a sub - data evaluation model, and output an evaluation reference value through the sub - data evaluation model according to the first limit rate, the second limit rate, and the first change rate; A third judgment module configured to judge whether the target corresponding to the first sub - data corresponding to the maximum evaluation reference value is a heating equipment or a pipeline. If it is a pipeline, send a control signal to keep the control valve of the target pipeline in the current state and increase the heating power of the heating equipment until the output power of the molten salt energy storage system reaches the second power demand. If it is a heating equipment, send a control signal to keep the current state of the heating equipment and an alarm signal; If the first change rate is not greater than zero, normally output the control signal of the control valve, pump or heating equipment corresponding to reaching the second power demand; A main control device, the main control device is connected to the power prediction module, the demand change module, the system simulation module, the first judgment module, the second judgment module and the third judgment module, and is used to execute the above-mentioned control method for the molten salt energy storage system.

[0025] The technical solution of the present invention has at least the following advantages and beneficial effects: The method provided by the present invention mainly includes judging whether the first change rate is greater than zero. If it is greater than zero, the first limit rate of each first sub-data is obtained according to the first sub-data and the corresponding limit value, the second limit rate of the first sub-data of the pipeline or heating equipment adjacent to the target first sub-data is obtained, and a sub-data evaluation model is established. The evaluation reference value is output through the sub-data evaluation model according to the first limit rate, the second limit rate and the first change rate. The higher the evaluation reference value, the more likely the current target is to have risks. The pipeline or heating equipment corresponding to the current highest evaluation reference value is obtained through the above method, and an alarm signal is sent to the remote control end. The personnel of the remote control end decide whether to continue to adjust the control valve or heating equipment normally according to the predicted power demand. In this way, the method can identify potential risks to a certain extent and feedback the results, so as to reduce the probability of risk as much as possible and make the system run stably and efficiently. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for use in the embodiments are briefly introduced below. It should be understood that the following drawings only show certain embodiments of the present invention and therefore should not be regarded as limiting the scope. For ordinary technicians in this field, other related drawings can be obtained based on these drawings without creative work.

[0027] Figure 1 It is a schematic diagram of the process of the present invention; Figure 2 It is a schematic diagram of the control system structure of the present invention. DETAILED DESCRIPTION

[0028] In order to make the purpose, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Generally, the components of the embodiments of the present invention described and shown in the drawings here can be arranged and designed in various different configurations.

[0029] The terms "first", "second", etc. in the description, claims and above-mentioned drawings of this application are used to distinguish similar objects and do not necessarily describe a specific order or sequence. The naming or numbering of steps in this application does not mean that the steps in the method flow must be executed in the time / logical sequence indicated by the naming or numbering. The named or numbered process steps can be changed in the execution order according to the technical purpose to be achieved, as long as the same or similar technical effects can be achieved.

[0030] Please refer to Figure 1 - Figure 2 , a control method for a molten salt energy storage system provided by the present invention includes: S101: Obtain the historical power demand data of the energy consumption end within the target period, preprocess the historical power demand data of the energy consumption end, establish a power prediction model, and output the power demand prediction result within a target period through the preprocessed historical power demand data of the energy consumption end and the power prediction model; Through the prediction of the power demand of the energy consumption end in the present invention, energy companies can arrange power generation and distribution plans more effectively, avoid waste of resources, and the prediction can help power suppliers identify high-demand periods in advance, so as to ensure that the power consumption needs of users can be met during peak periods and reduce the risk of power outages.

[0031] Accurate demand prediction helps to better integrate renewable energy sources such as wind energy and solar energy, because the power generation of such energy sources is affected by weather, and demand prediction can help balance supply and demand. Through reasonable prediction of demand, optimized use of electricity can be achieved, thereby reducing unnecessary energy waste and achieving the goals of energy conservation and emission reduction.

[0032] S102: Set an adjustment period, divide the target period into several adjustment time periods through the adjustment period, obtain the first power demand of the target adjustment time period, and obtain the predicted second power demand of the next judgment adjustment time period adjacent to the target adjustment time period through the power demand prediction result, and obtain the first change rate of the power demand through the first power demand and the second power demand; In this embodiment, the target period is set to 1 day, and the adjustment period is 1 hour or 0.5 hour, that is, 1 day is divided into 24 or 48 adjustment time periods. The prediction method provided in this solution is used for prediction in the previous adjustment time period, and at the beginning of an adjustment time period, the control command is executed.

[0033] Among them, combined with the prediction result of the previous step, the power demand of the energy consumption end in the next time period equivalent to the current time period can be obtained, so as to provide a basis for judgment and prediction for the subsequent steps.

[0034] S103: Establish a simulation model, and through the simulation model, obtain first sub-data of predicted pipeline detection data and heating equipment data in several segments according to the second power demand, and respectively obtain the limit values corresponding to several types of first sub-data; S104: Determine whether there is second sub-data greater than or equal to the limit value among several first sub-data. If so, obtain the pipeline or heating equipment corresponding to the second sub-data, send a control signal to keep the current state to the control valve or heating equipment on the pipeline corresponding to the second sub-data, and simultaneously send an alarm signal to the remote control terminal; In this embodiment, the pipeline detection data and heating equipment data of several segments include various index data such as pressure, temperature, and liquid flow rate. Each pipeline has its own standard, that is, the limit value that the pipeline can withstand. Based on the power demand of the energy-consuming end in the next adjustment time period, the various data that the pipeline needs to withstand are initially simulated and judged, and a preliminary judgment is made on whether any pipeline or heating equipment exceeds the limit value that the current pipeline or heating equipment can withstand when adjusting to the power demand of the energy-consuming end in the next adjustment time period. If so, a control signal is sent to the current pipeline or heating equipment in a timely manner, that is, to keep the current state, and the control valve or heating equipment is controlled not to make any changes. The remote control terminal makes the final decision on whether to continue. In this way, some obvious potential faults can be initially judged without wasting more computing power.

[0035] S105: If not, determine whether the first change rate is greater than zero. If it is greater than zero, obtain the first limit rate of each type of first sub-data according to the first sub-data and the corresponding limit value, obtain the second limit rate of the first sub-data of the pipeline or heating equipment adjacent to the target first sub-data, establish a sub-data evaluation model, and output an evaluation reference value through the sub-data evaluation model according to the first limit rate, the second limit rate, and the first change rate; In this step, the main purpose is to judge potential risks under normal circumstances to avoid losses. It mainly reflects the relationship between the data borne by the current pipeline or heating equipment and the limit value of the current pipeline or heating equipment. The closer to the limit value, the greater the potential risk may be. The evaluation reference value is calculated through the above first limit rate and second limit rate, and sorted to obtain the pipeline or heating equipment corresponding to the largest evaluation reference value, and key attention is paid.

[0036] S106: Determine whether the target corresponding to the first sub-data corresponding to the largest evaluation reference value is a heating equipment or a pipeline. If it is a pipeline, send a control signal to keep the control valve of the target pipeline in the current state and increase the heating power of the heating equipment until the output power of the molten salt energy storage system reaches the second power demand. If it is a heating equipment, send a control signal to keep the heating equipment in the current state and an alarm signal; S107: If the first change rate is not greater than zero, a control signal of a control valve, a pump or a heating device corresponding to the second power demand is normally output.

[0037] Secondly, if the first change rate is greater than zero, it means that the power demand of the energy-consuming end in the next adjustment time period is actually increasing, and it is necessary to perform a prediction as provided in the above scheme. If it is not greater than zero, it means that it is decreasing, and the pressure of each pipeline or heating equipment is decreasing or unchanged, so try to save the computing power of the system or reduce congestion. Among them, the first change rate can be the power demand of the energy-consuming end in the previous adjustment time period minus the power demand of the energy-consuming end in the next adjustment time period divided by the power demand of the energy-consuming end in the previous adjustment time period.

[0038] The method provided by the present invention mainly includes judging whether the first change rate is greater than zero. If it is greater than zero, the first limit rate of each first sub-data is obtained according to the first sub-data and the corresponding limit value, the second limit rate of the first sub-data of the pipeline or heating equipment adjacent to the target first sub-data is obtained, and a sub-data evaluation model is established. The evaluation reference value is output through the sub-data evaluation model according to the first limit rate, the second limit rate and the first change rate. The higher the evaluation reference value, the more likely the current target is to have risks. The pipeline or heating equipment corresponding to the current highest evaluation reference value is obtained through the above method, and an alarm signal is sent to the remote control end. The personnel of the remote control end decide whether to continue to adjust the control valve or heating equipment normally according to the predicted power demand. In this way, the method can identify potential risks to a certain extent and feedback the results, so as to reduce the probability of risk as much as possible and make the system run stably and efficiently.

[0039] In an exemplary embodiment of the present invention, establishing a power prediction model comprises: S201: Recording historical energy-consuming end power demand data as a load power sequence, fitting the historical energy-consuming end power demand data in the load power sequence to obtain a fitting function of the load growth rate;

[0040] In the formula, Represents the load power sequence, Indicates from the 1st to the Historical electricity load data, Indicates the number of historical data collected.

[0041]

[0042] In the formula, is the fitting function, , , are all fitting constants, is the coefficient of determination, is the natural constant.

[0043] S202: By correcting the growth cycle of the electricity load with the electricity-dependence residual value, the periodic characteristics of the electricity load growth are obtained; The growth trends fitted from historical electricity load data include three types: exponential growth, logarithmic growth, and linear growth.

[0044] Considering that the growth of the electricity load is affected by certain dependencies, in this study, based on the fitting function, the electricity-dependence residual value is calculated to correct the growth cycle of the electricity load.

[0045] S203: Define the LSTM memory unit in the model. The LSTM memory unit includes a forget gate, an input gate, and an output gate. Based on the basic information of the electricity load, three influencing factors are constructed, and the three influencing factors are input into the LSTM memory unit through the input gate, and the memory result is output through the analysis and memory of the LSTM memory unit at the output gate; According to the analyzed periodicity of the electricity load growth, the application of the LSTM+Attention model is introduced to track and identify the characteristics of the influencing factors of the electricity load in the region. According to the actual situation of the electricity load growth, this study uses the LSTM+Attention model to identify three dimensions: seasonal influencing factors, regional electricity characteristic influencing factors, and volatility influencing factors.

[0046] The above three influencing factors are input into the LSTM memory unit through the input gate of the model, and the memory result is output through the analysis and memory at the output gate.

[0047] S204: Add the Attention mechanism to construct a target model based on the Attention mechanism and the LSTM memory unit, and analyze the probability feature vector of the attention distribution of the output value of the LSTM memory unit in the model through the Attention mechanism.

[0048] The influence result of the influencing factor on the electricity load is obtained through the target model. Through the above analysis, the analysis result of the LSTM+Attention model on the influencing factors of the electricity load is obtained. According to this result, the tracking analysis of the load influence data in the region based on the LSTM+Attention model is realized, and the corresponding characteristics of the influencing factors of the electricity load are identified.

[0049]

[0050] In the formula, is the probability feature vector of the attention distribution, output value of the memory unit, is the weight matrix, is the Attention distribution; is the model state value, is one of the three influencing factors.

[0051] S205: Construct a linear regression equation for the characteristics of the influencing factors of electricity load in the region, construct a scaling matrix for predicting the growth of electricity load according to the linear regression equation, and perform scaling correction on the scaling matrix. Based on the corrected result, solve the predicted growth value of the electricity load.

[0052] Combined with the periodic change characteristics of the historical electricity load data obtained from the above analysis, this study uses the characteristics of the influencing factors of electricity load output by the LSTM+Attention model to predict the multi-dimensional influencing factors of the regional electricity load in the future for a period of time.

[0053] The linear regression equation includes:

[0054] In the formula, is the linear regression equation, and are two regression coefficients respectively, is the tracking and recognition value of the influencing factor by the LSTM+Attention model, is the variable parameter of the influencing factor, is the electricity load density variable.

[0055] In an exemplary embodiment of the present invention, the growth cycle of the electricity load is corrected by the residual value of electricity dependence, and the periodic characteristics of the growth of the electricity load include:

[0056]

[0057] In the formula, is the residual value of electricity dependence, is the type of growth trend fitting function, is the function value of the growth trend fitting function, is the electricity dependence coefficient, is the mean value of the function value within the prediction period, is the total number of types of the growth trend fitting function, is the growth cycle of the electricity load, is the growth cycle, is the coefficient of determination, is the fitting function.

[0058] An exemplary embodiment of the present invention, the three influencing factors include:

[0059]

[0060]

[0061] Wherein, is the first influencing factor of the electricity consumption load, is the second influencing factor of the electricity consumption load, is the third influencing factor of the electricity consumption load, is the real-time temperature of the current season, is the average temperature of the current season, is the seasonal fluctuation coefficient, is the number of seasons, is the seasonal parameter, is the power supply load margin, deviation cancellation coefficient, is the maximum difference in electricity consumption load, is the relative residual error, is the daily average load, is the valley electricity coefficient, is the daily maximum peak-valley difference, is the daily load fluctuation coefficient.

[0062] An exemplary embodiment of the present invention, solving the predicted growth value of the electricity consumption load includes:

[0063]

[0064] Wherein, is the predicted growth value of the electricity consumption load, is the power supply capacity, is the scaling matrix, is the area of the region of the load to be predicted, is the historical peak load value, is the time series parameter, is the scaling factor, where the subscript meaning of is the position in the matrix, is the time window length, that is, several hours in a cycle, is the linear regression equation, is a natural number, specifically the maximum value of the number of rows and columns of the position of the scaling factor in the matrix.

[0065] The scaling factor is determined by the linear derivative of the load density variable in the gear linear equation. The matrix is ​​used to scale and correct the range of regional load growth affected by periodic changes.

[0066] In an exemplary embodiment of the present invention, establishing a simulation model comprises: A plurality of historical detection data of pipelines and heating equipment as the power demand of the energy-consuming end increases are obtained respectively, and the detection data of the same type are divided into the same data group, and the relationship between the detection data and the power of the energy-consuming end is obtained according to the data group; The current power demand of the energy-consuming end is obtained, and the first sub-data corresponding to the prediction of the current power demand of the energy-consuming end is output through the data group.

[0067] Similarly, this step can also be performed by setting a fitting equation to find the relationship between each data type and the power of the energy-consuming end in a large amount of data, and then obtain the first sub-data corresponding to the prediction of the current power demand of the energy-consuming end. The present invention will not describe it in detail.

[0068] In an exemplary embodiment of the present invention, sending the control signal includes: Obtain the current system load rate of the molten salt energy storage system, set the system load alarm threshold, and determine whether the current system load rate reaches the system load alarm threshold; If it is not reached, the control signal is sent normally. If it is reached, the information content to be sent by the control signal is obtained. If the control signal is control information to maintain the current state, the maximum evaluation reference value is obtained, and a system load rate control model is established. Based on the maximum evaluation reference value, the system load rate that needs to be reduced is output through the system load rate control model, and a control signal is sent. If the control signal is output normally, the control signal is sent normally.

[0069] In this embodiment, since the control signal or alarm signal sent is relatively urgent, when the communication transmission channel of the system is congested, it causes a lag in the control signal or alarm signal and cannot fully exert its effect. Therefore, in this embodiment, if the system load rate reaches the system load alarm threshold, causing system congestion, a specific percentage of the reduced system load rate is given in combination with the current maximum evaluation reference value to minimize the possibility of delay in sending the control signal or alarm signal.

[0070] In an exemplary embodiment of the present invention, the system load rate control model includes:

[0071] In the formula, To reduce the system load rate, is the maximum evaluation reference value, is the number of all evaluation reference values, is the th evaluation reference value. In an exemplary embodiment of the present invention, the sub-data evaluation model includes:

[0072]

[0073]

[0074] In the formula, is the evaluation reference value, is the first change rate, is the first limit rate, is the second limit rate, , , …, are respectively the first limit rates of the first sub-data of the pipelines or heating devices adjacent to the target first sub-data from the 1st to the th first sub-data of the pipelines or heating devices adjacent to the target first sub-data, is the value of the first sub-data, is the limit value corresponding to the first sub-data.

[0075] The present invention also provides a control system for a molten salt energy storage system, including: A power prediction module configured to output a power demand prediction result within a target period through preprocessed historical power demand data of the energy consumption end and a power prediction model; A demand change module configured to obtain a first change rate of the power demand through a first power demand and a second power demand; A system simulation module configured to establish a simulation model and simulate to obtain first sub-data of predicted pipeline detection data of several segments and data of several heating devices; A first judgment module configured to judge whether there is a second sub-data greater than or equal to the limit value among several first sub-data. If so, obtain the pipeline or heating device corresponding to the second sub-data, send a control signal to maintain the current state to the control valve or heating device on the pipeline corresponding to the second sub-data, and simultaneously send an alarm signal to the remote control terminal; A second judgment module configured to, if not, judge whether the first change rate is greater than zero. If it is greater than zero, obtain the first limit rate of each first sub-data according to the first sub-data and the corresponding limit value, obtain the second limit rate of the first sub-data of the pipeline or heating device adjacent to the target first sub-data, establish a sub-data evaluation model, and output an evaluation reference value through the sub-data evaluation model according to the first limit rate, the second limit rate and the first change rate; A third judgment module, configured to judge whether the target corresponding to the first sub-data corresponding to the maximum evaluation reference value is a heating device or a pipeline. If it is a pipeline, a control signal is sent to keep the control valve of the target pipeline in the current state, and the heating device increases the heating power until the output power of the molten salt energy storage system reaches the second power demand. If it is a heating device, a control signal to keep the heating device in the current state and an alarm signal are sent. If the first change rate is not greater than zero, the control signal corresponding to the control valve, pump or heating device that reaches the second power demand is normally output; A main control device, which is connected to the power prediction module, the demand change module, the system simulation module, the first judgment module, the second judgment module and the third judgment module, and is used to execute the above control method for the molten salt energy storage system In addition, in each embodiment of the present invention, each functional unit can be integrated in a processing unit, or each unit can exist physically alone, or two or more units can be integrated in one unit. The above integrated unit can be implemented in the form of hardware or in the form of a software functional unit.

[0076] If the integrated unit is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. This computer software product is stored in a storage medium and includes several instructions to enable a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROMs), random access memories (RAMs), magnetic disks, or optical discs that can store program codes.

[0077] The above are only the preferred embodiments of the present invention and are not used to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A control method for a molten salt energy storage system, characterized in that: include: Establish a power prediction model, and output the power demand prediction result within a target period through the pre-processed historical power demand data of energy users and the power prediction model; Divide the target cycle into a plurality of adjustment time periods through the adjustment cycle, obtain a first power demand of the target adjustment time period, obtain a second power demand predicted for the next judgment adjustment time period adjacent to the target adjustment time period through the power demand prediction result, and obtain a first change rate of the power demand through the first power demand and the second power demand; Establish a simulation model, and obtain predicted first sub-data of several sections of pipeline detection data and several heating equipment data through the simulation model according to the second power demand, and respectively obtain limit values ​​corresponding to the first sub-data; Determine whether there is second sub-data greater than or equal to the limit value in the first sub-data, and if so, obtain the pipeline or heating device corresponding to the second sub-data, and send a control signal to maintain the current state to the control valve or heating device on the pipeline corresponding to the second sub-data; If not, when the first change rate is greater than zero, the first limit rate of each first sub-data is obtained according to the first sub-data and the corresponding limit value, the second limit rate of the first sub-data of the pipeline or heating equipment adjacent to the target first sub-data is obtained, a sub-data evaluation model is established, and an evaluation reference value is output through the sub-data evaluation model; Determine whether the target corresponding to the first sub-data corresponding to the maximum evaluation reference value is a heating device or a pipeline. If it is a pipeline, send a control signal that the control valve of the target pipeline maintains the current state and the heating device increases the heating power until the output power of the molten salt energy storage system reaches the second power requirement. If it is a heating device, send a control signal that the heating device maintains the current state; When the first change rate is not greater than zero, a control signal that meets the second power requirement is output normally.

2. The control method for a molten salt energy storage system according to claim 1, characterized in that: The power prediction model is established as follows: Recording the historical power demand data of the energy-consuming end as a load power sequence, fitting the historical power demand data of the energy-consuming end in the load power sequence to obtain a fitting function of the load growth rate; The growth cycle of the power load is corrected by the residual value of the power consumption dependence, so as to obtain the periodic characteristics of the power load growth; Define an LSTM memory unit in the model, wherein the LSTM memory unit includes a forget gate, an input gate, and an output gate, construct three influencing factors based on basic information of the power load, input the three influencing factors into the LSTM memory unit through the input gate, and output the memory result through the analysis memory of the LSTM memory unit at the output gate; Add the Attention mechanism, build a target model based on the Attention mechanism and LSTM memory unit, analyze the probability feature vector of the attention distribution of the output value of the LSTM memory unit in the model through the Attention mechanism, and obtain the impact of the influencing factors on the power load through the target model; A linear regression equation of the characteristics of the influencing factors of electricity load in the region is constructed, a scaling matrix for predicting the growth of electricity load is constructed based on the linear regression equation, and the scaling matrix is ​​scaled and corrected. Based on the corrected result, the predicted growth value of the electricity load is solved.

3. The control method for a molten salt energy storage system according to claim 2, characterized in that: The method of correcting the growth cycle of the power load by using the residual value of the power consumption dependency to obtain the periodic characteristics of the power load growth includes: In the formula, is the residual value of electricity dependence, is the growth trend fitting function type, is the function value of the growth trend fitting function, is the electricity dependence coefficient, is the mean value of the function in the forecast period, is the total number of types of growth trend fitting functions, is the electricity load growth cycle, For the growth cycle, is the coefficient of determination, is the fitting function.

4. The control method for a molten salt energy storage system according to claim 3, characterized in that: The three influencing factors include: In the formula, It is the first influencing factor of power load. is the second influencing factor of power load, It is the third influencing factor of power load. is the real-time temperature of the current season, is the mean temperature of the current season, is the seasonal fluctuation coefficient, is the number of seasons, is the seasonal parameter, is the power supply load margin, Deviation compensation coefficient, is the maximum power load difference, is the relative residual error, is the daily average load, is the valley current coefficient, is the maximum peak-to-valley difference of the day, is the daily load fluctuation coefficient.

5. The control method for a molten salt energy storage system according to claim 4, characterized in that: The method for solving the predicted growth value of the power load includes: In the formula, To predict the growth value of electricity load, is the power supply capacity, is the scaling matrix, is the area of ​​the load to be predicted, is the historical peak load value, is the timing parameter, is the time window length.

6. The control method for a molten salt energy storage system according to claim 5, characterized in that: The simulation model establishment comprises: A plurality of historical detection data of pipelines and heating equipment as the power demand of the energy-consuming end increases are obtained respectively, and the detection data of the same type are divided into the same data group, and the relationship between the detection data and the power of the energy-consuming end is obtained according to the data group; The current power demand of the energy-consuming end is obtained, and the first sub-data corresponding to the prediction of the current power demand of the energy-consuming end is output through the data group.

7. The control method for a molten salt energy storage system according to claim 6, characterized in that: Sending the control signal comprises: Obtain the current system load rate of the molten salt energy storage system, set the system load alarm threshold, and determine whether the current system load rate reaches the system load alarm threshold; If it is not reached, the control signal is sent normally. If it is reached, the information content to be sent by the control signal is obtained. If the control signal is control information to maintain the current state, the maximum evaluation reference value is obtained, and a system load rate control model is established. Based on the maximum evaluation reference value, the system load rate that needs to be reduced is output through the system load rate control model, and a control signal is sent. If the control signal is output normally, the control signal is sent normally.

8. The control method for a molten salt energy storage system according to claim 7, characterized in that: The system load rate control model includes: In the formula, To reduce the system load rate, is the maximum evaluation reference value, is the number of all evaluation reference values, For the An evaluation reference value.

9. The control method for a molten salt energy storage system according to claim 8, characterized in that: The sub-data evaluation model includes: In the formula, For evaluation reference value, is the first rate of change, is the first limiting rate, is the second limiting rate, , , …, The first limit rate of the first sub-data of the first pipeline or heating device adjacent to the first sub-data of the target is respectively The first limit rate of the first sub-data of the pipeline or heating equipment adjacent to the target first sub-data, is the value of the first sub-data, is the limit value corresponding to the first sub-data.

10. A control system for a molten salt energy storage system, characterized in that: include: A power prediction module is configured to output a power demand prediction result within a target period through preprocessed historical energy-consuming end power demand data and a power prediction model; A demand change module, configured to obtain a first change rate of power demand through the first power demand and the second power demand; A system simulation module is configured to establish a simulation model to simulate and obtain the predicted first sub-data of a plurality of pipeline detection data and a plurality of heating equipment data; A first judgment module is configured to judge whether there is second sub-data greater than or equal to a limit value among the plurality of first sub-data, and if so, obtain the pipeline or heating device corresponding to the second sub-data, send a control signal to maintain the current state to the control valve or heating device on the pipeline corresponding to the second sub-data, and send an alarm signal to the remote control terminal at the same time; The second judgment module is configured to judge whether the first change rate is greater than zero if not, and if greater than zero, obtain the first limit rate of each first sub-data according to the first sub-data and the corresponding limit value, obtain the second limit rate of the first sub-data of the pipeline or heating equipment adjacent to the target first sub-data, establish a sub-data evaluation model, and output an evaluation reference value through the sub-data evaluation model according to the first limit rate, the second limit rate and the first change rate; The third judgment module is configured to judge whether the target corresponding to the first sub-data corresponding to the maximum evaluation reference value is a heating device or a pipeline. If it is a pipeline, a control signal is issued to keep the control valve of the target pipeline in the current state and the heating device increases the heating power until the output power of the molten salt energy storage system reaches the second power requirement. If it is a heating device, a control signal and an alarm signal are issued to keep the heating device in the current state; if the first change rate is not greater than zero, a control signal of the control valve, pump or heating device corresponding to the second power requirement is normally output; A main control device, the main control device is connected to the power prediction module, the demand change module, the system simulation module, the first judgment module, the second judgment module and the third judgment module, and is used to execute the control method for the molten salt energy storage system described in any one of claims 1-9.

Images