Because of the complexity of construction projects and the frequency of project transactions, smart project management has a huge amount of transaction and operation data. Massive transaction and operation data urgently need a safe, efficient and low–cost storage and data analysis to realize fine management of buildings, low–cost transaction and interconnection of project information. Therefore, the construction project management system based on internet of things blockchain has been gradually improved, and more and more researches have been made on it. This article proposes a parallel design scheme. Parallel algorithms disperse the repeated calculation work on each node, reduce the calculation load and calculation time of a single node, effectively improve the efficiency of the algorithm, and enable it to process large–scale data of different sizes in parallel. Improve the algorithm according to the experiment, and analyze the performance of the parallel algorithm according to the experimental results. Experiments show that Apriori algorithm and K‐means algorithm based on MapReduce are more than 85° optimized than traditional data mining.

The electricity market is a complex nonlinear system, which not only follows the basic rules of market economy, but also is affected by the power system. This paper considers the non-stationarity of electricity futures and spot electricity price, and puts forward a forecasting model of futures electricity price by means of a cointegration theory to carry out an empirical research on the dynamic relationship between the electric spot markets and future markets. The research results show that there is a long-term equilibrium cointegration relationship between the spot electricity price and futures electricity price, and the combination of long-term relationship and short-term dynamic in time series of electricity price is realized, and a higher forecasting accuracy is obtained.