Since the image clarity of power equipment is easily disturbed by the external environment, the correct rate of recognition of power equipment by image recognition technology is easily affected. Based on this, this paper proposes a multimodal data fusion model for power equipment recognition. The Transformer model is selected to extract text features and image features separately, and then the features are fused and classified using a least squares support vector machine (LSSVM). Based on the data in the data space, the experimental part of this paper was completed. The experiments prove that the model in this paper has a greater advantage in the recognition of electrical equipment types, and the multimodal electrical equipment recognition results are better than the unimodal electrical equipment recognition results.

Data space is a technology system that allows data to be connected securely and efficiently. In the full life cycle of data space data, data collection, data storage, processing, transmission, exchange and destruction, and provision of services according to the dynamic changes of the subject's needs, is a brand-new data management model. This article first starts from the data security risk assessment of the data space of the electric power manufacturing industry, has analyzed the data characteristics of the data space of the electric power manufacturing industry, and sorted out the actual needs of data security. Then, based on the data life cycle process, the risk index of data space data security of electric power manufacturing enterprises are screened out. And used the method of combining Likert five-level scale and questionnaire to assess the amount of risk, graded risk index and proposed corresponding management measures, and constructed a dynamic cycle data security risk assessment model. Finally, a security governance system for data collection, data fusion and data transmission and other data full life cycle multi-faceted collaborative governance was established, and the data security risk of a Beijing electric power manufacturing company was evaluated, and put forward suggestions on how to coordinate the management of data security risks of power manufacturing enterprises.

The manufacturing industry is currently in a critical period of intelligent change, and the generation of massive amounts of data makes data management and data integration increasingly important. With the continuous upgrading of data management technology, how to handle diversified data and effectively collect multi-source heterogeneous data while ensuring data security has become the key to intelligent data management in current manufacturing enterprises. This paper analyzes the factors influencing the supply value chain in multi-value chain synergy, taking the external supply value chain of an electric power manufacturing company as an example. The gray correlation method is used to sort out the factors. Then the к-means method is used for data mining and cleaning. A dataset of key factors affecting the external value chain is established, and a data integration architecture for the dataspace of power manufacturing enterprises is constructed and empirically analyzed. The research results show that the data integration architecture can effectively tap into the management potential of power manufacturing enterprises in the external supply value chain and provide information solutions for the operation and management of power manufacturing enterprises.

The arrival of data space marks that all kinds of data applications can be recorded, stored and continuously expanded. The changes of data's source, volume and type have been increasing the difficulty of statistical analysis undoubtedly. In order to cope with the high-dimensional characteristics of data in data space, this paper discusses the challenges brought by high-dimensional data and high-dimensional models to traditional methods. The development of sparse modeling, the role of selection mechanism and the theoretical nature of penalty function method have also been combed in this paper. Finally, as an application, this paper discusses the feasibility of using high-dimensional sparse vector autoregressive model (HDS-VAR) to predict the profitability of manufacturing enterprises represented by electrical machinery and equipment manufacturing enterprises under the synergy of manufacturing value chain and service value chain.

Affected by the COVID-19, the global manufacturing industry has been greatly impacted. In order to adapt to the current new normal of economy, the multi-value chain collaborative operation mode of power manufacturing industry has come into being. In order to deeply study the influencing factors of multi-value chain collaborative operation efficiency in power manufacturing industry, this paper constructs an influencing factors system in terms of management level, technology level and policy level, combines fuzzy interpretative structural model (FISM) with analytic network process (ANP) to develop an analysis model from both qualitative and quantitative perspectives. Accordingly, it is suggested that: power manufacturing enterprises should promote the construction of R&D-production-sales-logistics-services multi-chain collaboration; promote the construction of data space to realize the sharing of data and information; accelerate the development of digital operation mode under Industry 4.0; and build third-party platform to efficiently integrate upstream and downstream resources.

A new idea is proposed that preprocessing is the key to improving the precision of short-term load forecasting (STLF). This paper presents a new model of STLF which is using support vector regression (SVR) based on pattern-base. Our model can be described as follows: firstly, it recognizes the different patterns of daily load according such features as weather and date type by means of data mining technology of classification and regression tree (CART); secondly, it sets up pattern-bases which are composed of daily load data sequence with highly similar features; thirdly, it establishes SVR forecasting model based on the pattern-base which matches to the forecasting day. Since the patterns of daily load are treated beforehand, the rule of the historical data sequence is more obvious. The model has many advantages: first, since the training data has similar pattern to the forecasting day, the model reflects the rule of daily load accurately and improves forecasting precision accordingly; second, as the pattern variables need not to be input into model, the mapping of the categorical variables is solved; third, as inputs are reduced, the model is simplified and the runtime is lessened. The simulation indicates that the new method is feasible and the forecasting precision is greatly improved.

Accurate forecasting of electricity load has been one of the most important issues in the electricity industry. Recently, along with power system privatization and deregulation, accurate forecast of electricity load has received increasing attention. According to the chaotic and non-linear characters of power load data, the model of support vector machines (SVM) based on Lyapunov exponents was established. The time series matrix was established according to the theory of phase-space reconstruction, and then Lyapunov exponents was computed to determine time delay and embedding dimension. Then support vector machines algorithm was used to predict power load. In order to prove the rationality of chosen dimension, another two random dimensions were selected to compare with the calculated dimension. And to prove the effectiveness of the model, BP algorithm was used to compare with the result of SVM. The results show that the model is effective and highly accurate in the forecasting of short-term power load. It is denoted that the model combining SVM and chaotic time series learning system has advantage than other models.