Power prediction method of integrated energy system, storage medium and terminal equipment
By building simulation models and gradient enhancement algorithm regression model, the interference and accuracy of power prediction data in the comprehensive energy system are solved, more accurate power prediction and energy management are achieved, and energy consumption costs are reduced.
Patent Information
- Application Number
- CN202410576840.6
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-05-10
- Publication Date
- 2024-11-08
AI Technical Summary
In the prior art, when predicting power in integrated energy systems, excessive data interference and poor prediction accuracy are caused by inaccurate system management strategies and increasing energy consumption costs.
By determining the index parameters according to the device type, building simulation models and gradient enhancement algorithm regression models, using historical monitoring data and simulation power data to train the final power prediction model, reducing data interference and improving prediction accuracy.
It realizes more accurate power prediction, reduces the system's hot backup capacity, optimizes energy management, reduces energy consumption costs, and improves the effectiveness of energy management solutions.
Smart Images

Figure CN118917164A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of integrated energy system control, and in particular to a power prediction method for an integrated energy system. Background Art
[0002] As the main mode of current energy system, the integrated energy system, an intelligent management system can gradually achieve multi-energy complementarity and meet the diversified energy needs of end users. In the field of energy consumption, with the widespread application of photovoltaic power generation and energy storage systems, the energy control process of intelligent integrated energy systems has become more and more complicated.
[0003] Accurate power prediction of integrated energy systems can overcome the impact of the randomness and volatility of distributed energy on the energy system. On the basis of ensuring the safe, stable and uninterrupted power supply of the system, it can provide data support for the formulation of reasonable energy scheduling and management strategies, thereby reducing the system's hot standby capacity and reducing energy costs, thus forming a better energy management solution.
[0004] The existing technology mainly uses historical monitoring data for prediction. The historical monitoring data collected is diverse, and the introduced data is not highly correlated with the prediction, resulting in increased interference. The existing method of building a model based on a neural network and then making predictions has the problem of poor prediction accuracy.
[0005] Therefore, how to improve the problems of excessive interference in data collected by existing technologies and poor prediction accuracy is a technical problem that needs to be solved urgently in this field. Summary of the invention
[0006] Based on this, the purpose of this application is to provide a power prediction method, medium and device for an integrated energy system to solve at least one technical problem mentioned in the above background technology.
[0007] In a first aspect, the present application provides a power prediction method for an integrated energy system, comprising:
[0008] According to the equipment type, determine the indicator parameters of each equipment type to obtain the historical monitoring data of each type of equipment;
[0009] Construct a simulation model with historical monitoring data as input and simulated power data as output; input the historical monitoring data of various types of equipment into the simulation model to obtain simulated power data of various types of equipment;
[0010] Based on the regression model of the gradient boosting algorithm, a power prediction model is constructed with historical monitoring data as input and predicted power data as output; the historical monitoring data is used as the training set and the simulated power data is used as the test verification set to continuously train the regression model and obtain the final power prediction model;
[0011] Obtain the historical monitoring data of the current device and input it into the final power prediction model to obtain its predicted power data.
[0012] Furthermore, it is characterized in that the device type includes:
[0013] One or more of power supply equipment, power-consuming equipment, heating equipment, heat-consuming equipment, cooling equipment, cooling equipment, gas supply equipment, and gas-consuming equipment.
[0014] Furthermore, the indicator parameters for each device type are:
[0015] The power supply / power-consuming equipment is the active power, the heating equipment is the thermal power, the heat-consuming equipment is the heat load power, the cooling equipment is the cooling power, the cooling equipment is the cooling load power, the gas supply equipment is the gas supply flow, and the gas-consuming equipment is the gas consumption flow.
[0016] Furthermore, the steps of obtaining historical monitoring data of various types of equipment include:
[0017] The average value of the detection data of each type of equipment per hour is collected, and the historical monitoring data of each type of equipment is obtained through comprehensive statistics.
[0018] Further, the step of constructing a simulation model with historical monitoring data as input and simulated power data as output includes:
[0019] Establish node power balance equation 2-1, 2-2 is the simulation model:
[0020] P i -P Gi +P Li =0 2-1
[0021] Q i -Q Gi +Q Li =0 2-2
[0022] Among them, P i and Q i Represent the external input active power and reactive power of the integrated energy system, P Gi and Q Gi Represents the active power and reactive power of the generator in the integrated energy system, P Li and Q Li It represents the load active power and reactive power of the integrated energy system.
[0023] Furthermore, the step of obtaining the simulated power data of each device also includes:
[0024] According to the balance relationship between the power input, output and environmental parameters of the integrated energy system, the node power balance equation 2-3 is optimized and established, and 2-4 is the simulation model:
[0025] P i -P Gi +P Li *E(x)=0 2-3
[0026] Q i -Q Gi +Q Li *E(x)=0 2-4
[0027] Among them, P i and Q i Represent the external input active power and reactive power of the integrated energy system, P Gi and Q Gi Represents the active power and reactive power of the generator in the integrated energy system, P Li and Q Li It represents the load active power and reactive power of the integrated energy system, and E(x) is the environmental parameter.
[0028] Furthermore, in the regression model, the form of the gradient boosting algorithm is shown in Formula 4-1:
[0029]
[0030] Where M is the number of numbers, T(x; θ m ) represents a decision tree, θ m The parameters of the decision tree are set. The regression model based on the gradient boosting algorithm is used to construct the next round of decision trees by fitting the difference between the historical monitoring data and the simulated power data, thereby gradually reducing the prediction error of the decision tree for the training data. When the training end condition is reached, the training is stopped to obtain the final power prediction model.
[0031] Further, including:
[0032] Display the historical monitoring data and predicted power data of each device on the pre-selected visualization chart.
[0033] In a second aspect, the present application further provides a computer storage medium storing an executable program code; the executable program code is used to execute the power prediction method for the integrated energy system described in any one of the first aspects.
[0034] In a third aspect, the present application also provides a terminal device comprising a memory and a processor; the memory stores program code executable by the processor; the program code is used to execute the power prediction method for the integrated energy system described in any one of the first aspects. BRIEF DESCRIPTION OF THE DRAWINGS
[0035] Figure 1 The present invention is a flowchart of an embodiment of a method for power prediction of an integrated energy system. DETAILED DESCRIPTION
[0036] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0037] It should be noted that if the embodiments of the present invention involve directional indications, such as up, down, left, right, front, back, etc., then the directional indication is only used to explain the relative position relationship, movement status, etc. between the components in a certain specific posture. If the specific posture changes, the directional indication will also change accordingly. In addition, if the embodiments of the present invention involve descriptions of "first, second", "S1, S2", "step one, step two", etc., then such descriptions are only used for descriptive purposes and cannot be understood as indicating or implying their relative importance or implicitly indicating the number of indicated technical features or indicating the execution order of the method, etc. Those skilled in the art can understand that anything that does not violate the gist of the invention under the technical concept of the invention should be included in the protection scope of the present invention.
[0038] like Figure 1 As shown, the present invention provides a power prediction method for an integrated energy system:
[0039] S1: According to the equipment type, determine the indicator parameters of each equipment type to obtain the historical monitoring data of each type of equipment;
[0040] Specifically, the integrated energy system preferably includes several types of equipment, and the corresponding indicator parameters are determined according to the type of each equipment. Equipment data is collected at set intervals and / or the average value of equipment data within the time period is obtained at set intervals to obtain historical monitoring data of each type of equipment; the set time is arbitrarily set by technical personnel in this field based on factors such as the specific equipment type, process conditions, and production environment.
[0041] For example, the comprehensive energy system is divided into eight types of equipment, namely power supply, power consumption, heating supply, heat consumption, cooling supply, cooling consumption, gas supply, and gas consumption. Corresponding indicator parameters are set according to the equipment types: active power for power supply / power consumption equipment, thermal power for heating equipment, thermal load power for heat consumption equipment, cooling power for cooling equipment, cooling load power for cooling equipment, gas supply flow for gas supply equipment, and gas consumption flow for gas consumption equipment. The average value of the detection data of each type of equipment per hour is collected, and comprehensive statistics are made to obtain the historical monitoring data of each type of equipment.
[0042] S2: construct a simulation model with historical monitoring data as input and simulated power data as output; input the historical monitoring data of various types of equipment into the simulation model to obtain simulated power data of various types of equipment;
[0043] Specifically, it is optional but not limited to obtaining parameters such as equipment type, energy flow direction and energy topology structure of each device in the integrated energy system, building a simulation model according to a pre-selected simulation software, and then inputting the obtained historical monitoring data into the simulation model for calculation to obtain the simulated power data of each device;
[0044] Preferably, the simulation software may be selected from but not limited to common simulation software such as MATLAB / Simulink, PSCAD / EMTDC, DIgSILENTPowerFactory, etc.
[0045] More preferably, taking the power supply / power consumption equipment in the integrated energy system as an example, it is optional but not limited to establishing the node power balance equation 2-1 and 2-2 as a simulation model according to the balance relationship between the power input and output of the integrated energy system:
[0046] P i -P Gi +P Li =0 2-1
[0047] Q i -Q Gi +Q Li =0 2-2
[0048] Among them, P i and Q i Represent the external input active power and reactive power of the integrated energy system, P Gi and Q Gi Represents the active power and reactive power of the generator in the integrated energy system, P Li and Q Li It represents the load active power and reactive power of the integrated energy system.
[0049] More preferably, since the production environment may interfere with the equipment, the influence of environmental parameters is preferably considered, and the node power balance equations 2-3 and 2-4 are optimized and established according to the balance relationship between the power input, output and environmental parameters of the integrated energy system as a simulation model:
[0050] P i -P Gi +P Li *E(x)=0 2-3
[0051] Q i -Q Gi +Q Li *E(x)=0 2-4
[0052] Among them, E(x) is the environmental parameter.
[0053] S3: Based on the regression model of the gradient boosting algorithm, a power prediction model is constructed with historical monitoring data as input and predicted power data as output; the historical monitoring data is used as the training set and the simulated power data is used as the test verification set to continuously train the regression model and obtain the final power prediction model;
[0054] Specifically, based on the specific form of the regression model, a multi-output single-output power prediction model is set, and then the historical monitoring data collected in step S1 is used as a training set, and the simulated power data obtained by simulation in step S2 is used as a test verification set. The training set is input into the regression model to obtain predicted data, and then the predicted data is verified according to the test verification set. The training end condition is set, and the regression model is continuously trained until the training end condition is reached, thereby constructing a final prediction model with historical monitoring data as input and predicted power data as output.
[0055] For example, in the regression model, the gradient boosting algorithm can be optionally expressed as 3-1:
[0056]
[0057] Where M is the number of numbers, T(x; θ m ) represents a decision tree, θ m The parameters of the decision tree are set, and the regression model based on the gradient boosting algorithm is used to construct the next round of decision trees by fitting the difference between the historical monitoring data and the simulated power data, thereby gradually reducing the prediction error of the decision tree for the training data until the iteration termination condition is reached.
[0058] Preferably, the iteration termination condition is optional but not limited to: when the maximum preset training times are reached, the training is stopped to obtain the final power prediction model; or when the performance of the strategy tree model meets certain conditions, such as the prediction error is not greater than the set error threshold, the training is stopped, at which time the error of the test prediction data will be in the optimal state; or when the training data set is used up, etc.
[0059] S4: Obtain historical monitoring data of the current device, input it into the final power prediction model, and obtain its predicted power data.
[0060] Specifically, according to the equipment type of the current equipment, historical monitoring data of corresponding indicator parameters are collected, and then input into the final power prediction model. After model calculation, the predicted power data of the current equipment can be obtained.
[0061] In this embodiment, a power prediction method for an integrated energy system of the present invention is provided, wherein step S1, according to the equipment type, various types of historical monitoring data are obtained to enrich the energy data source and expand the general applicability of the prediction result; step S2, since the historical monitoring data has a large randomness, a simulation model for a certain equipment type is constructed, and all types of historical monitoring data are input into the simulation model to obtain the simulation power data of the specific type to reduce the interference of other types of equipment data and expand the data characteristics of the specific type; step S3, based on the regression model of the gradient boosting algorithm, a power prediction model is constructed with the historical monitoring data as input and the predicted power data as output. The power prediction model uses the historical monitoring data obtained in step S1 as a training set on the one hand, and the simulated power data obtained in step S2 as a verification set on the other hand to train the regression model, so that the accuracy of the trained power prediction model is greatly improved, and the accuracy of the entire prediction result is increased. It has both universal applicability of all types and specific exclusivity of a certain type, which can fully utilize the richness of comprehensive energy, realize multi-energy complementarity and meet the diversified energy needs of end users; it can also avoid the problems of increased interference caused by the low correlation between the introduced data and the prediction direction due to the diversity of the collected historical monitoring data, so that the model prediction results are more in line with the current equipment type. On this basis, the existing technology cannot deal with the problems of data loss in the original data sample set, incomplete sampling of the time axis, incomplete sampling of the equipment, etc., and combined with simulation, it can solve the above problems and enhance the applicability of the scene. Step S4, the historical monitoring data of the current equipment is input into the final power prediction model, and its predicted power data is more accurate and more in line with reality.
[0062] Further preferably, after obtaining the predicted power data of each device, the present invention further includes:
[0063] S5: Display the historical monitoring data and predicted power data of each device on a pre-selected visualization chart.
[0064] Specifically, it is optional but not limited to technical personnel in this field to select a visualization chart based on the historical monitoring data and predicted power data of each device. Since the curve chart supports time axis translation, dragging, and zooming operations and is suitable for point-by-point drawing of the original data, it is preferred to display the historical monitoring data and predicted power data of each device on the curve chart, and draw the historical monitoring data as a solid line and the predicted power data as a dotted line to distinguish different data, thereby more intuitively showing the comparison between the data of the same device in the same time range; preferably, the visualization chart includes but is not limited to curve charts, bar charts, pie charts and other graphics commonly used to display statistical data.
[0065] On the other hand, the present invention also provides a computer storage medium storing executable program code; the executable program code is used to execute any of the above-mentioned power prediction methods for the integrated energy system.
[0066] On the other hand, the present invention also provides a terminal device, including a memory and a processor; the memory stores program code that can be executed by the processor; the program code is used to execute any of the above-mentioned power prediction methods for the integrated energy system.
[0067] Exemplarily, the program code may be divided into one or more modules / units, which are stored in the memory and executed by the processor to complete the present invention. The one or more modules / units may be a series of computer program instruction segments capable of completing specific functions, which are used to describe the execution process of the program code in the terminal device.
[0068] The terminal device may be a computing device such as a desktop computer, a notebook, a PDA, a cloud server, etc. The terminal device may include, but is not limited to, a processor and a memory. Those skilled in the art will appreciate that the terminal device may also include an input / output device, a network access device, a bus, etc.
[0069] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), field-programmable gate arrays (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor, etc.
[0070] The memory may be an internal storage unit of the terminal device, such as a hard disk or a memory. The memory may also be an external storage device of the terminal device, such as a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), etc. equipped on the terminal device. Further, the memory may also include both an internal storage unit of the terminal device and an external storage device. The memory is used to store the program code and other programs and data required by the terminal device. The memory may also be used to temporarily store data that has been output or is to be output.
[0071] The above-mentioned computer storage medium and terminal device are created based on the power prediction method of the above-mentioned integrated energy system. Their technical functions and beneficial effects are not repeated here. The various technical features of the above-mentioned embodiments can be combined arbitrarily. In order to make the description concise, all possible combinations of the various technical features in the above-mentioned embodiments are not described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.
[0072] The above-mentioned embodiments only express several implementation methods of the present invention, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the invention patent. It should be pointed out that, for ordinary technicians in this field, several variations and improvements can be made without departing from the concept of the present invention, and these all belong to the protection scope of the present invention. Therefore, the protection scope of the patent of the present invention shall be subject to the attached claims.
Claims
1. A power prediction method for an integrated energy system, It is characterized in that include: According to the equipment type, determine the indicator parameters of each equipment type to obtain the historical monitoring data of each type of equipment; Construct a simulation model with historical monitoring data as input and simulated power data as output; input the historical monitoring data of various types of equipment into the simulation model to obtain simulated power data of various types of equipment; Based on the regression model of the gradient boosting algorithm, a power prediction model is constructed with historical monitoring data as input and predicted power data as output; the historical monitoring data is used as the training set and the simulated power data is used as the test verification set to continuously train the regression model and obtain the final power prediction model; Obtain the historical monitoring data of the current device and input it into the final power prediction model to obtain its predicted power data.
2. The method according to claim 1, It is characterized in that Device types include: One or more of power supply equipment, power-consuming equipment, heating equipment, heat-consuming equipment, cooling equipment, cooling equipment, gas supply equipment, and gas-consuming equipment.
3. The method according to claim 2, It is characterized in that The indicator parameters for each device type are: The power supply / power-consuming equipment is the active power, the heating equipment is the thermal power, the heat-consuming equipment is the heat load power, the cooling equipment is the cooling power, the cooling equipment is the cooling load power, the gas supply equipment is the gas supply flow, and the gas-consuming equipment is the gas consumption flow.
4. The method according to claim 1, It is characterized in that The steps to obtain historical monitoring data for each type of equipment include: The average value of the detection data of each type of equipment per hour is collected, and the historical monitoring data of each type of equipment is obtained through comprehensive statistics.
5. The method according to claim 1, It is characterized in that The steps of constructing a simulation model with historical monitoring data as input and simulated power data as output include: Establish node power balance equation 2-1, 2-2 is the simulation model: P i -P Gi +P Li =0 2-1 Q i -Q Gi +Q Li =0 2-2 Among them, P i and Q i Represent the external input active power and reactive power of the integrated energy system, P Gi and Q Gi Represents the active power and reactive power of the generator in the integrated energy system, P Li and Q Li It represents the load active power and reactive power of the integrated energy system.
6. The method according to claim 2, It is characterized in that The step of obtaining the simulated power data of each device also includes: According to the balance relationship between the power input, output and environmental parameters of the integrated energy system, the node power balance equation 2-3 is optimized and established, and 2-4 is the simulation model: P i -P Gi +P Li *E(x)=0 2-3 Q i -Q Gi +Q Li *E(x)=0 2-4 Among them, P i and Q i Represent the external input active power and reactive power of the integrated energy system, P Gi and Q Gi Represents the active power and reactive power of the generator in the integrated energy system, P Li and Q Li It represents the load active power and reactive power of the integrated energy system, and E(x) is the environmental parameter.
7. The method according to claim 1, It is characterized in that In the regression model, the form of the gradient boosting algorithm is shown in Formula 4-1: Where M is the number of numbers, T(x; θ m ) represents a decision tree, θ m The parameters of the decision tree are set. The regression model based on the gradient boosting algorithm is used to construct the next round of decision trees by fitting the difference between the historical monitoring data and the simulated power data, thereby gradually reducing the prediction error of the decision tree for the training data. When the training end condition is reached, the training is stopped to obtain the final power prediction model.
8. The method according to claim 1, It is characterized in that include: Display the historical monitoring data and predicted power data of each device on the pre-selected visualization chart.
9. A computer storage medium, It is characterized in that An executable program code is stored; the executable program code is used to execute the power prediction method for the integrated energy system described in any one of claims 1-8.
10. A terminal device, It is characterized in that It comprises a memory and a processor; the memory stores program codes executable by the processor; the program codes are used to execute the power prediction method of the integrated energy system as described in any one of claims 1-8.
