Although some existing collaborative intrusion detection schemes can increase the detection performance by dynamically allocating detection resources in smart grids, these related works fail to consider the optimization of resource allocation between IDSs under the condition of resource restriction. In this paper, considering the effect of resource restriction, we propose a resource allocation scheme for two-layer collaborative IDSs based on sharing strategies in smart grids. In the first layer of our scheme, we model the interaction between the IDSs and the attackers through a stochastic game based on sharing strategies, where we provide each IDS with two different options for its strategy updating at each stage in the stochastic game. Then the resource updating strategies of the IDSs are obtained through this proposed model. Further, in the second layer we quantify the effect of detection resource restriction, and we propose a resource allocation method under the condition of detection resource restriction, where each IDS can obtain its detection resources according to the results generated by our proposed stochastic game. Based on our experimental analysis, compared with other resource allocation schemes, our proposed scheme can more quickly achieve the Nash equilibrium between the IDSs and the attackers to make the IDSs obtain more rewards, and then can more rationally promote the IDSs to update their detection resources so that the IDSs obtain the optimal detection strategies under the condition of resource restriction. Our proposed scheme can achieve effective detection resource allocation between IDSs for the security of neighborhood area network in smart grids.