A power grid cost analysis method and system based on artificial intelligence
Through the grid cost analysis method based on artificial intelligence, the equipment status evaluation model and cost prediction model are used to quickly and accurately calculate the operating cost of power grid power generation equipment, solving the problem of low efficiency in power grid cost analysis in the existing technology, and achieving efficient and accurate Cost analysis.
Patent Information
- Application Number
- CN202411132024.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-08-19
- Publication Date
- 2024-11-05
- Estimated Expiration
- 2044-08-19
AI Technical Summary
In the prior art, the cost analysis of power grid power generation equipment is low, and manual data processing is difficult to process quickly and accurately.
Using the artificial intelligence-based grid cost analysis method, by obtaining the operating data, equipment type and rated power of the power generation equipment, the equipment status evaluation model, normalization processing, grade score and weight relationship, combined with the cost prediction model, quickly and accurately Calculate the operating cost of the power grid.
It improves the efficiency and accuracy of cost analysis of power grid power generation equipment, avoids inefficiency and possible errors in manual data processing, and saves manpower and resources.
Smart Images

Figure CN118657546B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of power grid cost assessment, and in particular to a power grid cost analysis method and system based on artificial intelligence. Background Art
[0002] With the continuous advancement of science and technology, the demand for electric energy is also increasing. Electricity has become the basis of production and life in modern society. Almost all industries and fields require electricity support. Whether it is industrial production, transportation or daily life, electricity plays a vital role, which makes the scale of the power grid continue to increase and decrease.
[0003] Cost analysis of power generation equipment in the power grid is a very important task. It can better understand the operating costs of power generation equipment and formulate corresponding energy-saving and consumption-reducing measures. However, conventional cost analysis methods require manual data processing and analysis. Therefore, when processing large amounts of data, manual work may not be able to quickly and accurately complete the cost analysis of power generation equipment in the power grid. Summary of the invention
[0004] The problem solved by the present invention is how to improve the efficiency of cost analysis of power grid power generation equipment.
[0005] In order to solve the above problems, the present invention provides a power grid cost analysis method and system based on artificial intelligence.
[0006] In a first aspect, the present invention provides a power grid cost analysis method based on artificial intelligence, comprising:
[0007] Obtain the operating data, equipment type and rated power generation of each power generation equipment in the power grid;
[0008] Input the operating data into a preset equipment status evaluation model and output corresponding equipment status indicators;
[0009] According to the rated power generation corresponding to all the power generation equipment, a rated power generation coefficient corresponding to each power generation equipment is obtained through normalization processing;
[0010] Classify all the power generation equipment according to the equipment type to obtain a grade score for each power generation equipment;
[0011] According to the rated power generation coefficient, the grade score and the equipment status index corresponding to the power generation equipment, the operation weight corresponding to the power generation equipment is obtained through a weight relationship;
[0012] According to the operation weights of all the power generation equipment and the corresponding operation data, the predicted operation cost of the power grid is obtained through a preset cost prediction model.
[0013] Optionally, obtaining the rated power generation coefficient corresponding to each power generation device by normalizing the rated power generation corresponding to all the power generation devices includes:
[0014] The total rated power generation is obtained by summing up all the rated power generation; and the total rated power generation is divided by the number of power generation equipment to obtain the average rated power generation of all the power generation equipment;
[0015] According to the rated power generation corresponding to each power generation device and the average rated power generation, the rated power generation coefficient corresponding to each power generation device is obtained through a preset normalization relationship;
[0016] The normalization relationship satisfies:
[0017] ;
[0018] Among them, Y j is the rated power generation coefficient corresponding to the j-th power generation equipment, X j is the rated power generation corresponding to the j-th power generation equipment, μ is the average rated power generation, and m is the number of the power generation equipment.
[0019] Optionally, classifying all the power generation equipment according to the equipment type to obtain a grade score for each power generation equipment includes:
[0020] Acquire a preset power generation equipment grade matching rule, wherein the power generation equipment grade matching rule includes a correspondence between the equipment type in the power grid and the grade score;
[0021] According to the equipment type corresponding to the power generation equipment, the grade score corresponding to the power generation equipment is determined through the corresponding relationship.
[0022] Optionally, before obtaining the preset power generation equipment level matching rule, the method further includes:
[0023] Sum the rated power generation of all the power generation devices of the same device type to obtain the total rated power generation of the device type;
[0024] According to the total rated power generation of the equipment type corresponding to each equipment type, the grade score corresponding to each equipment type is obtained through a grade score relationship;
[0025] The corresponding relationship is generated according to the device type and the corresponding grade score.
[0026] Optionally, the rating relationship satisfies:
[0027] ;
[0028] Among them, C k is the grade score corresponding to the kth device type, P k is the total rated power generation of the device type corresponding to the kth device type, and h is the number of device types in the power grid.
[0029] Optionally, obtaining the operation weight corresponding to the power generation equipment through a weight relationship according to the rated power generation coefficient corresponding to the power generation equipment, the grade score and the equipment status index includes:
[0030] Adding the rated power generation coefficient corresponding to the power generation equipment to the grade score to obtain the equipment weight of the power generation equipment;
[0031] The operation weight of the power generation equipment is obtained according to the product of the equipment weight corresponding to the power generation equipment and the equipment status index.
[0032] Optionally, the training process of the cost prediction model includes:
[0033] Obtaining historical operation data and historical operation costs corresponding to different time nodes of the power generation equipment;
[0034] Screening all the historical operation data through linear fitting to determine key historical operation data;
[0035] Generate a training data set based on all of the key historical operating data and the corresponding historical operating costs;
[0036] The preset initial model is trained using the training data set to obtain the trained cost prediction model.
[0037] Optionally, the screening of all the historical operation data by linear fitting to determine key historical operation data includes:
[0038] According to all the historical operation data, determining the fitting coefficient corresponding to each of the historical operation data through a fitting relationship;
[0039] When the fitting coefficient is less than a preset threshold, the corresponding historical operation data is determined as key historical operation data.
[0040] Optionally, the fitting relationship satisfies:
[0041] ;
[0042] Among them, b iis the fitting coefficient corresponding to the i-th historical operation data, a ij is the jth data type in the ith historical operating data, m is the number of the historical operating data, and n is the number of the data types in the historical operating data.
[0043] In a second aspect, the present invention provides a power grid cost analysis system based on artificial intelligence, comprising:
[0044] An acquisition module is used to obtain the operating data, equipment type and rated power generation corresponding to each power generation equipment in the power grid;
[0045] A status evaluation module, used to input the operating data into a preset equipment status evaluation model and output corresponding equipment status indicators;
[0046] A statistical module, used for obtaining a rated power coefficient corresponding to each power generation device through normalization processing according to the rated power generation corresponding to all the power generation devices;
[0047] A grade evaluation module, used to classify all the power generation equipment according to the equipment type, and obtain a grade score for each power generation equipment;
[0048] A weight evaluation module, used to obtain the operation weight corresponding to the power generation equipment through a weight relationship according to the rated power generation coefficient corresponding to the power generation equipment, the grade score and the equipment status index;
[0049] The cost prediction module is used to obtain the predicted operating cost of the power grid through a preset cost prediction model according to the operating weights of all the power generation equipment and the corresponding operating data.
[0050] The beneficial effects of the power grid cost analysis method and system based on artificial intelligence of the present invention are as follows: according to the acquired operation data, the operation status index of the equipment can be obtained through the pre-trained equipment status evaluation model, and the operation status of all power generation equipment in the power grid can be accurately judged through the operation status index, so that the power supply capacity of the power generation equipment to the power grid can be accurately judged. In addition, the corresponding rated power coefficient is obtained according to the power generation power of each power generation equipment, and the maximum power generation capacity of each power generation equipment can be obtained through the rated power coefficient, so that the importance of each power generation equipment to the power grid can be accurately evaluated. At the same time, the importance of different types of power generation equipment is scored, so as to obtain the grade score corresponding to the type of each power generation equipment, and the contribution of the power generation equipment to the overall power supply of the power grid can be further judged through the grade score. Then, according to the rated power coefficient, grade score and equipment status index corresponding to each power generation equipment, the operation weight of the power generation equipment is calculated, and the importance of the power generation equipment to the power supply capacity of the power grid is reflected as a whole through the operation weight. The operating weights and corresponding operating data of all power generation equipment are used to output the predicted operating cost of the power grid through a pre-trained cost prediction model, so that the operating cost of the power generation equipment in the power grid can be obtained quickly and accurately, avoiding the problem of inefficient manual processing and analysis of data, saving manpower and resource investment, and effectively improving the efficiency of power generation cost analysis of power generation equipment in the power grid. At the same time, it avoids the errors that may be caused by manual data processing and analysis, and improves the accuracy of cost analysis of power generation equipment in the power grid. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] Figure 1 A schematic diagram of a flow chart of a power grid cost analysis method based on artificial intelligence according to an embodiment of the present invention;
[0052] Figure 2 The present invention is a schematic diagram of the structure of an artificial intelligence-based power grid cost analysis system according to an embodiment of the present invention. DETAILED DESCRIPTION
[0053] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings. Although certain embodiments of the present invention are shown in the accompanying drawings, it should be understood that the present invention can be implemented in various forms and should not be interpreted as being limited to the embodiments described herein. On the contrary, these embodiments are provided to provide a more thorough and complete understanding of the present invention. It should be understood that the drawings and embodiments of the present invention are only for exemplary purposes and are not intended to limit the scope of protection of the present invention.
[0054] It should be understood that the various steps described in the method embodiments of the present invention may be performed in different orders and / or in parallel. In addition, the method embodiments may include additional steps and / or omit the steps shown. The scope of the present invention is not limited in this respect.
[0055] The term "including" and its variations used in this document are open inclusions, that is, "including but not limited to"; the term "based on" means "based at least in part on"; the term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one other embodiment"; the term "some embodiments" means "at least some embodiments"; the term "optionally" means "optional embodiments". The relevant definitions of other terms will be given in the following description. It should be noted that the concepts of "first", "second", etc. mentioned in the present invention are only used to distinguish different devices, modules or units, and are not used to limit the order or interdependence of the functions performed by these devices, modules or units.
[0056] It should be noted that the modifications of "one" and "plurality" mentioned in the present invention are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise clearly indicated in the context, it should be understood as "one or more".
[0057] The names of the messages or information exchanged between multiple devices in the embodiments of the present invention are only used for illustrative purposes, and are not used to limit the scope of these messages or information.
[0058] In view of the problems existing in the above-mentioned related technologies, this embodiment provides a power grid cost analysis method and system based on artificial intelligence.
[0059] like Figure 1 As shown, an embodiment of the present invention provides an artificial intelligence-based power grid cost analysis method, including:
[0060] S100, obtaining operation data, equipment type and rated power generation corresponding to each power generation equipment in the power grid.
[0061] Specifically, the operation data, type of power generation equipment and rated power generation of the power generation equipment corresponding to each access node in the power grid are obtained, wherein the types of power generation equipment include thermal power generation, hydropower generation, wind power generation and photovoltaic power generation, etc. The operation data include the voltage value, current value, output AC frequency, output power, power factor and equipment operating temperature of the power generation equipment output, which can reflect the data information of the operation status of the power generation equipment. The rated power generation refers to the maximum power generation of each power generation equipment under standard conditions. The state information of the power generation equipment can be accurately understood through the operation data of the power generation equipment. In addition, for different operation states of the equipment, the corresponding labor cost and maintenance cost are also relatively fixed. For example, when the power generation equipment is running at full load, the labor cost and maintenance cost invested are also the largest. Similarly, when the power generation equipment fails or the operation load is reduced according to the scheduling requirements, the corresponding operation and maintenance personnel and the maintenance cost invested will also be reduced accordingly. Therefore, the operation state of the equipment can also indirectly reflect the investment information of the maintenance cost of the power generation equipment. The operation data can provide rich and accurate data information for the cost analysis of the power generation equipment, thereby improving the accuracy of the cost analysis.
[0062] S200, inputting the operation data into a preset equipment status evaluation model, and outputting corresponding equipment status indicators.
[0063] Specifically, historical operation data groups of power generation equipment in different time periods are obtained, including historical operation data such as current, voltage, temperature, vibration, lubricating oil pressure and temperature, etc., the historical operation data are screened by data fitting, and abnormal historical operation data are deleted. A training data group is constructed according to each group of historical data groups after screening, and corresponding weight coefficients are pre-set according to the importance of different historical operation data in the training data group. The equipment status score corresponding to the historical operation data group in each time period is calculated through a preset equipment status relationship, that is, the equipment status index. The equipment status index can be used to accurately judge the operating status of the power generation equipment and promptly detect abnormalities of the power generation equipment. If the equipment status index deviates from the standard value, it means that the equipment may fail.
[0064] Exemplarily, the device status relationship satisfies:
[0065] ;
[0066] Among them, S is the device status score, a i is the i-th historical running data, b i is the weight coefficient corresponding to the i-th historical operation data, and n is the number of historical operation data.
[0067] Furthermore, the corresponding training data group is labeled according to the calculated equipment status score, and all the labeled training data groups constitute a model training set. The initial model is trained by the model training set to finally obtain a trained equipment status evaluation model, wherein the annotations of the training data group are used to compare with the status indicators output by the initial model according to the training data group through a preset loss function, and the weight coefficients of the initial model set are updated according to the comparison result. The initial model is iteratively trained with the training data set until the deviation between the output status indicator and the corresponding annotation meets the preset requirements, that is, the result of the loss function meets the preset requirements. At this time, the corresponding model is determined as the trained equipment status evaluation model. The loss function can adopt a squared loss function (SLF), a cross entropy loss function (CELF) or an absolute value loss function (ALF), etc. Selecting an appropriate loss function for different types of power generation equipment in the branch can help optimize the performance and generalization ability of the model.
[0068] Specifically, according to the acquired operating data of the power generation equipment, the operating data is input into a preset equipment status assessment model to obtain the equipment status index corresponding to each power generation equipment, that is, the status score of the power generation equipment. It should be noted that due to the large differences in the operating data of different types of power generation equipment, corresponding equipment status assessment models are pre-trained for different types of power generation equipment. For example, according to the historical operating data of thermal power generation equipment, a device status assessment model for thermal power generation equipment is trained; according to the historical operating data of hydropower generation equipment, a device status assessment model for hydropower generation equipment is trained; the acquired thermal power generation operating data is input into the thermal power generation equipment status assessment model, and the corresponding thermal power generation equipment status index is output; the acquired hydropower generation operating data is input into the hydropower generation status assessment model, and the corresponding hydropower generation equipment status index is output. Through the equipment status index assessment model, the operating status of the current power generation equipment can be quickly and accurately assessed, so that the power generation equipment status can be more accurately judged according to the operating status.
[0069] S300, obtaining a rated power coefficient corresponding to each power generation device through normalization processing according to the rated power generation corresponding to all the power generation devices.
[0070] Specifically, based on the obtained rated power generation of each power generation equipment, the rated power generation coefficient corresponding to each power generation equipment is determined through the normalization relationship. The power coefficient can accurately obtain the proportional relationship of the power generation capacity of each power generation equipment in the entire power grid, thereby judging the contribution of the power generation equipment to the power transmission of the power grid, that is, the importance of the power generation equipment. Among them, the normalization relationship includes normalization operation methods such as maximum-minimum normalization relationship or decimal normalization relationship. Any of the above normalization relationships can be used to normalize the rated power generation of all power generation equipment.
[0071] Specifically, the maximum-minimum normalization relationship satisfies:
[0072] ;
[0073] The decimal normalization relationship satisfies:
[0074] ;
[0075] Among them, Y j is the rated power generation coefficient corresponding to the j-th power generation equipment, X j is the rated power of the jth power generation equipment, X min is the minimum rated power among all rated power, X max is the maximum rated power generation of all power generation, and m is the number of rated power generation.
[0076] S400: Classify all the power generation equipment according to the equipment type to obtain a grade score for each power generation equipment.
[0077] Specifically, according to the equipment types corresponding to all power generation equipment in the power grid, the power generation equipment is graded according to their importance to obtain a grade score for each equipment type, thereby determining the grade score corresponding to each power generation equipment according to its equipment type.
[0078] Furthermore, since there may be multiple different types of power generation equipment in the power grid at the same time, and the importance of each type of power generation equipment in the power grid may be different, first, the total power generation of each power generation type, or the sum of all rated power generation of the same power generation type in the power grid, is counted, and the grade score of each power generation type is determined according to the proportion of the total rated power generation of each power generation type to the rated power generation of all power generation equipment in the power grid. The grade score can be used to accurately judge the ability and importance of each power generation type to transmit electric energy to the power grid. For example, in areas where hydropower generation is concentrated, the main power generation equipment in the power grid is hydropower generation equipment. At this time, the grade of hydropower generation equipment needs to be set to a higher level, and the grades of other auxiliary power generation equipment are relatively low. When thermal power generation is the main power generation equipment, the grade of thermal power generation equipment is set to a higher level, and the grades of other auxiliary power generation equipment are relatively low.
[0079] Furthermore, the type of equipment corresponding to each power generation equipment is determined
[0080] For example: the rating satisfies:
[0081] ;
[0082] Among them, C k is the grade score corresponding to the kth device type, P k is the sum of the rated powers of all power generation equipment corresponding to the kth equipment type, and h is the number of equipment types in the power grid, that is, the power generation equipment in the power grid can be divided into h different equipment types.
[0083] S500, obtaining an operation weight corresponding to the power generation equipment through a weight relationship according to the rated power generation coefficient, the grade score and the equipment status index corresponding to the power generation equipment.
[0084] Specifically, according to the rated power generation coefficient, grade score and equipment status index corresponding to each power generation equipment, the corresponding operating weight of each power generation equipment is calculated through the weight relationship. Through the power generation of the power generation equipment, the importance of the equipment type in the power grid and the operating status of the equipment, the importance of each power generation equipment to the power grid under the operating state is comprehensively calculated. Therefore, through the operating weights of different power generation equipment, the contribution of each power generation equipment to the power grid can be accurately evaluated.
[0085] S600, obtaining the predicted operating cost of the power grid through a preset cost prediction model according to the operating weights of all the power generation equipment and the corresponding operating data.
[0086] Specifically, based on the operating weights and operating data of all the power generation equipment in the power grid, the predicted operating costs of all the power generation equipment in the power grid are obtained through a cost prediction model that has been pre-trained based on historical data. The operating costs of the power generation equipment in the power grid can be quickly and accurately obtained through model prediction, avoiding the processing and analysis of a large amount of power generation equipment data, saving manpower and resources, and improving the efficiency of cost analysis.
[0087] In this embodiment, according to the acquired operation data, the operation status index of the equipment can be obtained through the pre-trained equipment status evaluation model. The operation status index can accurately determine the operation status of all power generation equipment in the power grid, so that the power supply capacity of the power generation equipment to the power grid can be accurately determined. In addition, the corresponding rated power generation coefficient is obtained according to the power generation power of each power generation equipment, and the maximum power generation capacity of each power generation equipment can be obtained through the rated power generation coefficient, so that the importance of each power generation equipment to the power grid can be accurately evaluated. At the same time, the importance of different types of power generation equipment is scored, so as to obtain the grade score corresponding to the type of each power generation equipment, and the contribution of the power generation equipment to the overall power supply of the power grid can be further determined through the grade score. Then, according to the rated power generation coefficient, grade score and equipment status index corresponding to each power generation equipment, the operation weight of the power generation equipment is calculated, and the importance of the power generation equipment to the power supply capacity of the power grid is reflected as a whole through the operation weight. The operating weights and corresponding operating data of all power generation equipment are used to output the predicted operating cost of the power grid through a pre-trained cost prediction model, so that the operating cost of the power generation equipment in the power grid can be obtained quickly and accurately, avoiding the problem of inefficient manual processing and analysis of data, saving manpower and resource investment, and effectively improving the efficiency of power generation cost analysis of power generation equipment in the power grid. At the same time, it avoids the errors that may be caused by manual data processing and analysis, and improves the accuracy of cost analysis of power generation equipment in the power grid.
[0088] Optionally, obtaining the rated power generation coefficient corresponding to each power generation device by normalizing the rated power generation corresponding to all the power generation devices includes:
[0089] The total rated power generation is obtained by summing up all the rated power generation; and the total rated power generation is divided by the number of power generation equipment to obtain the average rated power generation of all the power generation equipment;
[0090] According to the rated power generation corresponding to each power generation device and the average rated power generation, the rated power generation coefficient corresponding to each power generation device is obtained through a preset normalization relationship;
[0091] The normalization relationship satisfies:
[0092] ;
[0093] Among them, Y j is the rated power generation coefficient corresponding to the j-th power generation equipment, X j is the rated power generation corresponding to the j-th power generation equipment, μ is the average rated power generation, and m is the number of the power generation equipment.
[0094] Specifically, the total rated power generation power of the power grid is obtained by summing up the rated power generation power corresponding to all the power generation equipment in the power grid, and then the total rated power generation power is divided by the number of corresponding power generation equipment to obtain the average rated power generation power corresponding to all the power generation equipment. According to the rated power generation power of each power generation equipment and the average rated power generation power, the rated power generation coefficient corresponding to each power generation equipment is calculated through a pre-set normalized relationship.
[0095] In this optional embodiment, by normalizing the rated power generation of each power generation equipment, it can be ensured that the obtained rated power generation coefficient is within the same scale range, thereby avoiding errors in the data processing structure caused by the scale differences between the rated power generation of different power generation equipment. The rated power generation coefficient can be used to better compare the differences between the rated power generation of different power generation equipment, and then more accurately judge the differences in the amount of electric energy that each power generation equipment can provide to the power grid, thereby obtaining the importance of each power generation equipment to the power supply of the power grid.
[0096] Optionally, classifying all the power generation equipment according to the equipment type to obtain a grade score for each power generation equipment includes:
[0097] Acquire a preset power generation equipment grade matching rule, wherein the power generation equipment grade matching rule includes a correspondence between the equipment type in the power grid and the grade score;
[0098] According to the equipment type corresponding to the power generation equipment, the grade score corresponding to the power generation equipment is determined through the corresponding relationship.
[0099] Optionally, before obtaining the preset power generation equipment level matching rule, the method further includes:
[0100] Sum the rated power generation of all the power generation devices of the same device type to obtain the total rated power generation of the device type;
[0101] According to the total rated power generation of the equipment type corresponding to each equipment type, the grade score corresponding to each equipment type is obtained through a grade score relationship;
[0102] The corresponding relationship is generated according to the device type and the corresponding grade score.
[0103] Optionally, the rating relationship satisfies:
[0104] ;
[0105] Among them, C k is the grade score corresponding to the kth device type, P k is the total power of the device type corresponding to the kth device type, and h is the number of device types in the power grid.
[0106] Specifically, the importance of different types of power generation equipment in the power grid is obtained, and the levels of different equipment types are scored to obtain the correspondence between each equipment type and the grade score, and a power generation equipment grade matching rule is constructed based on all equipment levels and corresponding grade scores. According to the equipment type corresponding to each power generation equipment, the corresponding grade score is determined in the power generation equipment grade matching rule.
[0107] Further, the rated power generation of all power generation equipment of the same equipment type in the power grid is obtained, and the rated power generation of all the same equipment types is added together to obtain the corresponding total rated power generation of all the equipment types, that is, the total rated power generation of the equipment type, and the corresponding grade score of each equipment type is calculated according to the total rated power generation of the equipment type corresponding to all equipment types through the grade score relationship, that is, the proportion of the total rated power of the equipment type corresponding to each equipment type to the total rated power of all power generation equipment in the power grid, so as to generate the corresponding relationship between the equipment type and the grade score according to the equipment type and the corresponding grade score. For example, the equipment types in the power grid include thermal power generation, wind power generation and photovoltaic power generation, and the total rated power generation corresponding to all thermal power generation equipment, the total rated power generation corresponding to all thermal power generation equipment, and the total rated power generation corresponding to all photovoltaic power generation equipment are calculated respectively. The grade scores corresponding to the thermal power generation equipment, wind power generation equipment and photovoltaic power generation equipment are calculated respectively through the grade score relationship, and then the corresponding relationship is generated according to the above three power generation types and the corresponding grade scores, so as to construct the corresponding power generation equipment grade matching rules of the power grid through the corresponding relationship.
[0108] In this optional embodiment, the grade score corresponding to each power generation equipment is determined by the grade scores of different equipment types, so that the importance of each power generation equipment in the power grid and its contribution to the transmission of electric energy to the power grid can be accurately judged based on the grade scores corresponding to the power generation equipment.
[0109] Optionally, obtaining the operation weight corresponding to the power generation equipment through a weight relationship according to the rated power generation coefficient corresponding to the power generation equipment, the grade score and the equipment status index includes:
[0110] Adding the rated power generation coefficient corresponding to the power generation equipment to the grade score to obtain the equipment weight of the power generation equipment;
[0111] The operation weight of the power generation equipment is obtained according to the product of the equipment weight corresponding to the power generation equipment and the equipment status index.
[0112] Specifically, the rated power generation power and grade score corresponding to each power generation equipment are summed to obtain the equipment weight corresponding to each power generation equipment, and the equipment weight corresponding to each power generation equipment is multiplied by the corresponding grade score to obtain the operation weight corresponding to each power generation equipment. Therefore, the importance of the corresponding power generation equipment to the power grid under the operating state can be accurately judged through the operation weight of each power generation equipment, and the contribution of each power generation equipment to the power grid can be accurately evaluated through the operation weights of different power generation equipment.
[0113] Optionally, the training process of the cost prediction model includes:
[0114] Obtaining historical operation data and historical operation costs corresponding to different time nodes of the power generation equipment;
[0115] Screening all the historical operation data through linear fitting to determine key historical operation data;
[0116] Generate a training data set based on all of the key historical operating data and the corresponding historical operating costs;
[0117] The preset initial model is trained using the training data set to obtain the trained cost prediction model.
[0118] Specifically, the operating data of the power generation equipment at different time nodes of the entire power grid and the corresponding operating costs, that is, the historical operating data and the corresponding historical operating costs, are obtained, and screening is performed through fitting, that is, the same type of operating data at different time nodes are fitted to obtain the corresponding standard operating data, and screening is performed according to the deviation between each type of data and the corresponding standard operating data. When the difference between the historical operating data and the corresponding standard operating data exceeds a preset threshold, it is determined that the time node corresponding to the operating data does not meet the requirements and is deleted. Finally, the key historical operating data is screened, and a training data set is generated according to all the key historical operating data and the corresponding historical operating costs. The initial model is iteratively trained with the key historical operating data in the training data set, and the weights in the initial model are continuously updated according to the comparison results of the output value of the initial model and the corresponding historical operating costs. Finally, the model whose output value meets the requirements is determined as the trained cost prediction model.
[0119] In this optional embodiment, by training the initial model to obtain a trained cost prediction model, the operating cost of the power generation equipment in the power grid can be quickly predicted through the cost prediction model, and by inputting the operating data of all the power generation equipment in the power grid in combination with the corresponding operating weights into the cost prediction model, the corresponding operating cost of the entire power grid can be accurately output, wherein the operating weight can further reflect the importance of each discovered device to the power grid during its operation. Therefore, through the operating weight and corresponding operating data of the power generation equipment, the cost prediction model can more accurately output the prediction results, thereby improving the efficiency and accuracy of the cost analysis of the power generation equipment in the power grid.
[0120] Optionally, all the historical operation data are screened by linear fitting to determine key historical operation data, including:
[0121] According to all the historical operation data, determining the fitting coefficient corresponding to each of the historical operation data through a fitting relationship;
[0122] When the fitting coefficient is less than a preset threshold, the corresponding historical operation data is determined as key historical operation data.
[0123] Optionally, the fitting relationship satisfies:
[0124] ;
[0125] Among them, b i is the fitting coefficient corresponding to the i-th historical operation data, a ij is the jth data type in the ith historical operating data, m is the number of the historical operating data, and n is the number of the data types in the historical operating data.
[0126] Specifically, based on all historical operating data, the fitting coefficient corresponding to each historical operating data is calculated through the fitting relationship, and the fitting coefficient is compared with a preset threshold. When the operating data corresponding to the fitting coefficient is greater than or equal to the preset threshold, the operating data is deleted, and the historical operating data whose fitting coefficients are all less than the preset threshold are determined as key operating historical data.
[0127] Exemplarily, based on all the historical operating data, the fitting coefficient corresponding to each historical operating data is calculated by fitting relationship. For example, m historical operating data are selected, wherein each historical operating data includes n different data types. The same data type in all the historical operating data is summed and divided by m to obtain the average value corresponding to the data type. Thus, performing the above calculation on each data type can obtain the average value of n different data types, and taking the absolute value of the difference between each data type in the historical operating data and the average value corresponding to each data type to obtain the deviation value of each data type in the historical operating data, and summing the deviation values of all data types in the historical operating data to obtain the fitting coefficient corresponding to the historical operating data.
[0128] In this optional embodiment, the fitting coefficient of each historical operating data is calculated to obtain the difference between each data type in the operating history data and the corresponding data type in other historical operating data, and the operating history data with large differences are deleted to obtain key historical operating data that meets the requirements. Through the key operating history data, the accuracy of model training can be improved, so that the finally trained cost prediction model can output accurate operating costs based on the operating data.
[0129] like Figure 2 As shown, an embodiment of the present invention provides an artificial intelligence-based power grid cost analysis system, including:
[0130] An acquisition module is used to obtain the operating data, equipment type and rated power generation corresponding to each power generation equipment in the power grid;
[0131] A status evaluation module, used to input the operating data into a preset equipment status evaluation model and output corresponding equipment status indicators;
[0132] A statistical module, used for obtaining a rated power coefficient corresponding to each power generation device through normalization processing according to the rated power generation corresponding to all the power generation devices;
[0133] A grade evaluation module, used to classify all the power generation equipment according to the equipment type, and obtain a grade score for each power generation equipment;
[0134] A weight evaluation module, used to obtain the operation weight corresponding to the power generation equipment through a weight relationship according to the rated power generation coefficient corresponding to the power generation equipment, the grade score and the equipment status index;
[0135] The cost prediction module is used to obtain the predicted operating cost of the power grid through a preset cost prediction model based on the operating weights of all the power generation equipment and the corresponding operating data.
[0136] The artificial intelligence-based power grid cost analysis system of this embodiment is used to implement the artificial intelligence-based power grid cost analysis method as described above. Its advantages over the existing technology are the same as the advantages of the above-mentioned artificial intelligence-based power grid cost analysis method over the existing technology, and will not be repeated here.
[0137] A person of ordinary skill in the art can understand that all or part of the processes in the above-mentioned embodiment method can be implemented by instructing the relevant hardware through a computer program, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, the storage medium can be a disk, an optical disk, a read-only memory (ROM) or a random access memory (RAM), etc. In the present application, the unit described as a separate component may or may not be physically separated, and the component displayed as a unit may or may not be a physical unit, that is, it may be located in one place, or it may be distributed on multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the embodiment of the present invention. In addition, each functional unit in each embodiment of the present invention can be integrated in a processing unit, or each unit can exist physically separately, or two or more units can be integrated in one unit. The above-mentioned integrated unit can be implemented in the form of hardware or in the form of a software functional unit.
[0138] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.
Claims
1. A power grid cost analysis method based on artificial intelligence, characterized in that: include: Obtain the operating data, equipment type and rated power generation of each power generation equipment in the power grid; Input the operating data into a preset equipment status evaluation model and output corresponding equipment status indicators; According to the rated power generation corresponding to all the power generation equipment, a rated power generation coefficient corresponding to each power generation equipment is obtained through normalization processing; Classify all the power generation equipment according to the equipment type to obtain a grade score for each power generation equipment; According to the rated power generation coefficient, the grade score and the equipment status index corresponding to the power generation equipment, the operation weight corresponding to the power generation equipment is obtained through a weight relationship; According to the operation weights of all the power generation equipment and the corresponding operation data, a predicted operation cost of the power grid is obtained through a preset cost prediction model; The rated power generation coefficient corresponding to each power generation device is obtained by normalizing the rated power generation corresponding to all the power generation devices, including: The total rated power generation is obtained by summing up all the rated power generation; and the total rated power generation is divided by the number of power generation equipment to obtain the average rated power generation of all the power generation equipment; According to the rated power generation corresponding to each power generation device and the average rated power generation, the rated power generation coefficient corresponding to each power generation device is obtained through a preset normalization relationship; The normalization relationship satisfies: ; Among them, Y j is the rated power generation coefficient corresponding to the j-th generating equipment, X j is the rated power generation corresponding to the j-th power generation equipment, μ is the average rated power generation, and m is the number of the power generation equipment; The step of obtaining the operation weight corresponding to the power generation equipment through a weight relationship according to the rated power generation coefficient corresponding to the power generation equipment, the grade score and the equipment status index comprises: Adding the rated power generation coefficient corresponding to the power generation equipment to the grade score to obtain the equipment weight of the power generation equipment; Obtaining the operation weight of the power generation equipment according to the product of the equipment weight corresponding to the power generation equipment and the equipment status index; The training process of the cost prediction model includes: Obtaining historical operation data and historical operation costs corresponding to different time nodes of the power generation equipment; Screening all the historical operation data through linear fitting to determine key historical operation data; Generate a training data set based on all of the key historical operating data and the corresponding historical operating costs; Training the preset initial model using the training data set to obtain the trained cost prediction model; The linear fitting is used to screen all the historical operation data to determine key historical operation data, including: According to all the historical operation data, determining the fitting coefficient corresponding to each of the historical operation data through a fitting relationship; When the fitting coefficient is less than a preset threshold, the corresponding historical operation data is determined as key historical operation data; The fitting relationship satisfies: ; Among them, b i is the fitting coefficient corresponding to the i-th historical operation data, a ij is the jth data type in the ith historical operating data, m is the number of the historical operating data, and n is the number of the data types in the historical operating data.
2. The power grid cost analysis method based on artificial intelligence according to claim 1 is characterized in that: The step of classifying all the power generation equipment according to the equipment type to obtain a grade score for each power generation equipment includes: Acquire a preset power generation equipment grade matching rule, wherein the power generation equipment grade matching rule includes a correspondence between the equipment type in the power grid and the grade score; According to the equipment type corresponding to the power generation equipment, the grade score corresponding to the power generation equipment is determined through the corresponding relationship.
3. The power grid cost analysis method based on artificial intelligence according to claim 2 is characterized in that: Before obtaining the preset power generation equipment level matching rule, the method further includes: Sum the rated power generation of all the power generation devices of the same device type to obtain the total rated power generation of the device type; According to the total rated power generation of the equipment type corresponding to each equipment type, the grade score corresponding to each equipment type is obtained through a grade score relationship; The corresponding relationship is generated according to the device type and the corresponding grade score.
4. The power grid cost analysis method based on artificial intelligence according to claim 3 is characterized in that: The rating relationship satisfies: ; Among them, C k is the grade score corresponding to the kth device type, P k is the total rated power generation of the device type corresponding to the kth device type, and h is the number of device types in the power grid.
5. An artificial intelligence-based power grid cost analysis system, characterized in that: include: An acquisition module is used to obtain the operating data, equipment type and rated power generation corresponding to each power generation equipment in the power grid; A status evaluation module, used to input the operating data into a preset equipment status evaluation model and output corresponding equipment status indicators; A statistical module, used for obtaining a rated power coefficient corresponding to each power generation device through normalization processing according to the rated power generation corresponding to all the power generation devices; The rated power generation coefficient corresponding to each power generation device is obtained by normalization according to the rated power generation corresponding to all the power generation devices, including: obtaining the total rated power generation according to the sum of all the rated power generation; and obtaining the average rated power generation of all the power generation devices by dividing the total rated power generation by the number of power generation devices; obtaining the rated power generation coefficient corresponding to each power generation device by a preset normalization relationship according to the rated power generation corresponding to each power generation device and the average rated power generation; The normalization relationship satisfies: ; Among them, Y j is the rated power factor corresponding to the j-th power generation equipment, X j is the rated power generation corresponding to the j-th power generation equipment, μ is the average rated power generation, and m is the number of the power generation equipment; A grade evaluation module, used to classify all the power generation equipment according to the equipment type, and obtain a grade score for each power generation equipment; A weight evaluation module, used to obtain the operation weight corresponding to the power generation equipment through a weight relationship according to the rated power generation coefficient corresponding to the power generation equipment, the grade score and the equipment status index; Wherein, obtaining the operation weight corresponding to the power generation equipment through a weight relationship according to the rated power generation coefficient corresponding to the power generation equipment, the grade score and the equipment status index includes: adding the rated power generation coefficient corresponding to the power generation equipment and the grade score to obtain the equipment weight of the power generation equipment; obtaining the operation weight of the power generation equipment according to the product of the equipment weight corresponding to the power generation equipment and the equipment status index; A cost prediction module, used to obtain the predicted operating cost of the power grid through a preset cost prediction model according to the operating weights of all the power generation equipment and the corresponding operating data; The training process of the cost prediction model includes: obtaining the historical operation data and historical operation costs corresponding to different time nodes of the power generation equipment; screening all the historical operation data through linear fitting to determine the key historical operation data; generating a training data set according to all the key historical operation data and the corresponding historical operation costs; training the preset initial model through the training data set to obtain the trained cost prediction model; The method of screening all the historical operation data through linear fitting to determine the key historical operation data includes: determining a fitting coefficient corresponding to each of the historical operation data through a fitting relationship according to all the historical operation data; when the fitting coefficient is less than a preset threshold, determining the corresponding historical operation data as the key historical operation data; The fitting relationship satisfies: ; Among them, b i is the fitting coefficient corresponding to the i-th historical operation data, a ij is the jth data type in the ith historical operating data, m is the number of the historical operating data, and n is the number of the data types in the historical operating data.
