Because the basic characteristic of HSIC is to imitate the control expert's control behavior, its control algorithm is multi-modal control, and it is used interchangeably among many modal control. The algorithm can perfectly coordinate the many conflicting control quality requirements in the control system. For example, robustness and accuracy, speed and stability. Third, the characteristics of water supply pump station and its control requirements In the development of urban construction, intelligent building has become a standard for people to pursue good living conditions, and water supply pump station is an indispensable link in the intelligent building group, a reasonable choice of pump control , Not only can reduce project costs, but also energy. 1. Water supply system characteristics. For specific objects, the most prominent feature of the user's water is random, which user has water, how much water, when the water, etc., have great uncertainty. From a macro point of view, the characteristics of water supply system mainly in the following areas: (1) the unknown system parameters, time-varying, randomness and dispersion; (2) the system lag unknown and time-varying; (3) Serious non-linearity; (4) the correlation between system variables; and (5) the unknown, diversity and randomness of environmental interference. 2. Control problems. The above characteristics belong to the control problem of the complex object (or process) with uncertainty. The traditional control is powerless, mainly in the following aspects: (1) Uncertainty problem. Many of the control problems in water systems are uncertain, difficult to model using traditional methods, and therefore fail to achieve effective control. (2) highly nonlinear. There are a lot of nonlinear problems in the water supply system. In the traditional control theory, the nonlinear theory is much less mature than the linear theory, because the method is too complicated and difficult to apply. (3) semi-structured and unstructured issues. Traditional control theory can not solve the problem of semi-structured and unstructured water supply system. (4) Water supply system complexity issues. Complex relationships between the various subsystems intricate, highly coupled between the various elements, mutual restraint, the external environment is extremely complex, the traditional control of the lack of effective solutions. (5) reliability issues. The conventional mathematical model-based control problem tends to be an interdependent one, and the reliability of the control of simple systems is not conspicuous. For the water supply system, if the above method is used, the entire control system may collapse due to the change of conditions. Thus it can be seen that the traditional methods can not effectively control such systems and more effective control methods must be explored. 3. Control requirements. No matter what kind of control measures, we must meet the user's water needs (that is, to maintain a certain pressure), to protect the environment from noise pollution, in addition to consider energy saving. Therefore, the control requirements can be determined to meet the user's water supply requirements under the premise of minimizing environmental pollution and energy conservation. Fourth, the selection of control strategies Control strategy selection and the characteristics of the controlled object are closely related to the wrong or inappropriate control strategies often lead to very poor control, or even lead to out of control system. At present, in the modern water supply pump station in order to save energy are generally used in the inverter, to improve the quality of control to create a good condition. There are PID control modules in the inverter, but the PID algorithm is not suitable for the complicated system of water supply with uncertainty. Artificial neural networks, due to the limitations of the method, also have difficulty controlling the subject in question. The expert control system (ECS) is difficult to implement due to the collection of characteristic information, the expression of characteristic information and the establishment of a complete knowledge base. It is not always a good choice to adopt an expert control system. The fuzzy controller (FC) based on fuzzy control theory can realize the control of complex industrial processes. Its control of quality and effectiveness is still satisfactory, is an alternative strategy. Humanoid intelligent control, experts use HISC and FC control strategies to simulate the complex object (or process) with uncertainty respectively. Although both of them calculate control quantities based on error and error rate of change, HISC and FC can implement effective control, but the control quality and robustness of the former is better, so the use of HSIC to complete the uncertainty of the strong (strong coupling) characteristics of the field test, HISC and FC can implement effective control, The control of the water system is - a more sensible choice. V. CONCLUSION Intelligent control has been widely used in many fields such as industry, agriculture and military. It has solved a large number of practical control applications that can not be solved by traditional control and has shown great vitality and development prospects. With the development of basic theory and practice Application of the expansion of intelligent control will achieve a big leap in the field of control.